Sydney Harbour beach erosion – actions of the sea or stormwater runoff?

Thomas Mortlock, Barry Hanstrum and Mingzhu Wang

The heavy rain event on 28 November 2018 caused widespread flooding in the Illawarra, Sydney Metro and Central Coast areas, with some suburbs experiencing up to 61 mm in 30 minutes in Mosman, over 100 mm in two hours in the Sydney CBD and lower north shore, and 130 mm in 24 hours in Mosman. This type of springtime flash flooding is not uncommon, but is often associated with urban surface water flooding, and not coastal erosion.

From a coastal perspective, it raises an interesting question for erosion of Sydney’s North Shore and Middle Harbour beaches. These areas include some of Sydney’s most expensive residential and commercial property, and are usually regarded as protected, low wave energy beach locations with minimal erosion risk.

However – as demonstrated over the past few days – erosion can occur at these locations, and not necessarily due to waves from the seaward side but from urban stormwater runoff from the landward side (or to complicate things, a combination of both). The primary cause of erosion and consequential infrastructure damage can be important with regards to insurance against ‘actions of the sea’.

Rain then wind then waves

The 28 November rain event developed from a NW-SE oriented trough over the continent, with several low-pressure cells developing 1 – 2 days prior to 28 November (Figure 1, A – C). The deepest of these was centered over coastal NSW, reaching a moderate 994 hPa by the morning of 28 November (Figure 1, D).

The first front and associated rainfall (blue dotted line in Figure 1) passed over the Sydney Harbour area in the early morning of 28 November (6 to 8 AM). There was a maximum of 130 mm of rain in Mosman and 111 mm at The Rocks (Observatory Hill) – more than a month’s volume for these areas. A second front affected Western Sydney in the afternoon of 28 November (Figure 1 F) but did not significantly impact habour beach locations.

Winds were not a key feature of this event and peaked a few hours after the most intense rainfall was experienced on the North Shore – reaching just below 50 km/hr at 10:30 AM on 28 November (BoM Observatory Hill, BoM 2018b).

As the low-pressure cell moved offshore (Figure 1, G – H), rain and wind dropped while wave heights at the open coast began to rise (Figure 1). Observations at the Sydney wave buoy, located approximately 30 km off Narrabeen on Sydney’s Northern Beaches, show waves from the SE peaked in the late afternoon of 28 November, with maximum wave heights reaching between 7 and 10 m (Figure 2).

Figure 1. Synoptic evolution of the 28 November 2018 rain event in 6-hourly intervals from 11 AM on 27 November to 5 AM on 29 November. Images modified from BoM (2018a).
Figure 2. Wave height and direction observed at the Sydney wave buoy off Narrabeen through the 28 November rain event. Source: Manly Hydraulics Lab (2018).

A different type of East Coast Low

The 28 November event was an East Coast Low (ECL) – but not the type normally associated with coastal erosion. It developed from a continental / inland trough low, typical of a spring-time ECL (Browning and Goodwin, 2013), whereas most highly erosive ECLs occur in the winter months and have a more meridional (shore-parallel) track. The severe weather, while very intense was short lived because of the steady movement of the low.

The heavy rain was driven by intense thunderstorms that formed in the northwest of the Sydney Basin and tracked southeast. According to design rainfall depths, there is only a 1 % chance of this intensity of rainfall on the Lower North Shore (up to 61 mm in 30 minutes in Mosman) occurring in any one year (ARR, 2016). The intense rainfall was probably a result of cooler-than-normal land surface temperatures on the coast and relatively mild ocean temperatures, resulting in strong convection and rainfall (Browning pers. comms.).

This particular event had the characteristics of an extra-tropical cyclone and was driven by the dynamics of the upper atmosphere in a similar way to the mechanism for classic ECL development. Rain and wind were to the south of the centre of the low with clear sky and westerlies to the north.

It was a very unusual weather pattern for late November – the abnormal westerly flow to the north of the low was responsible for the record maximum temperature along the tropical QLD coast and catastrophic fire danger ratings. It’s very hard to get strong surface westerlies at these latitudes at this time of the year. The reason for this may be the unusual amount of Antarctic sea ice for this time of year (Goodwin, pers. comms.), which acts to displace the atmospheric cells in the Southern Hemisphere equatorward.

Causes of coastal erosion

In Sydney Harbour, the main cause of beach erosion was urban stormwater runoff (overland and drainage), not waves – an important distinction for insurance. Balmoral Beach in Mosman is a good example of the vulnerability of these largely protected, low wave energy coastal sites to this type of erosion. A steep, urbanized hinterland directly behind the beachfront led to high flow volumes running down several ‘urban rivers’ towards the beach (Figure 3). This washed out beach sand from the landward side, causing flooding and some significant erosion in places (Figure 4).

Figure 3. Conceptual plot of stormwater runoff-induced erosion at Balmoral Beach on 28 November 2018. Erosion hotspots correspond to lowest drainage points. Inset shows an example of an ‘urban river’ down a steep catchment in Neutral Bay.

It is possible for beaches in Middle Harbour to experience wave-driven coastal erosion. If wave heights outside Sydney Harbour are large, and the wave direction is from the SE, wind-waves can refract into Middle Harbour beaches and cause erosion. Interestingly, even beaches deeper in the harbour can experience erosion during ECLs without being directly exposed to the ocean wind waves. This (hypothetically – not tested) occurs through the propagation of infragravity wave energy – very long period, low amplitude waves that can bounce off headlands to reach otherwise protected locations. This type of wave energy is not detected by conventional wave buoys and is often associated with phenomena such as surf-beat or tsunamis. An example of this was during the April 2015 ECL, when Clontarf Beach – located deep inside Middle Harbour next to Spit Bridge – suffered severe erosion which may have been related to infragravity waves.

Figure 4. Erosion at Balmoral Beach caused by stormwater runoff. The beach level used to be at the height of the drainage covers in (A). Photos taken by author on 29/11/18.


Springtime ECLs are most commonly associated with flash flooding, and not coastal erosion because of the predominantly zonal track and limited wind and wave component of these events. 28 November, however, was a reminder that coastal erosion is not just caused by ‘actions of the sea’ but can result from a combination of heavy rainfall and a steep, urbanized hinterland. Fortunately, this type of rainfall is often localized – but it can lead to beach loss and infrastructure damage in some surprising places.

Risk Frontiers is currently building an East Coast Low loss model to assist insurers to price risk during these types of events.


Australian Rainfall Runoff [ARR] (2016), Design Rainfalls, available at, accessed 30/11/18.

Browning, S. and Goodwin, I.D. (2013), Large scale influences on the evolution of winter subtropical maritime cyclones affecting Australia’s east coast. Monthly Weather Review, 141, 2416-2431.

Bureau of Meteorology [BoM] (2018a), Latest colour mean sea-level pressure analysis, available at, accessed 29/11/18.

Bureau of Meteorology [BoM] (2018b), Latest weather observations for Sydney – Observatory Hill, available at, accessed 29/11/18.

Manly Hydraulics Lab [MHL] (2018), Sydney offshore wave height, period and direction, available at, accessed 29/11/18.


Key Conclusions of the U.S. National Climate Assessment 2018

Paul Somerville and Thomas Mortlock, Risk Frontiers

Figure 1 The US Government’s Fourth National Climate Assessment Report

The U.S. federal government on Friday 23 November released a long-awaited report (NCA4) with an unmistakable message: the effects of climate change, including deadly wildfires, increasingly debilitating hurricanes and heat waves, are already battering the United States, and the danger of more such catastrophes is worsening. The report’s authors, who represent numerous federal agencies, say they are more certain than ever that climate change poses a severe threat to Americans’ health and pocketbooks, as well as to the country’s infrastructure and natural resources. And while it avoids policy recommendations, the report’s sense of urgency and alarm stands in stark contrast to the lack of any apparent plan from President Trump to tackle the problems, which, according to the government he runs, are increasingly dire.

The congressionally mandated document, the first of its kind issued during the Trump administration, details how climate-fueled disasters and other types of changes are becoming more commonplace throughout the country and how much worse they could become in the absence of efforts to combat global warming.

The Fourth National Climate Assessment (NCA4) contains two volumes. Volume II draws on the founda­tional science described in Volume I, the Cli­mate Science Special Report (CSSR). Volume II focuses on the human welfare, societal, and environmental elements of climate change and variability for 10 regions and 18 national top­ics, with particular attention paid to observed and projected risks, impacts, consideration of risk reduction, and implications under dif­ferent mitigation pathways. Where possible, NCA4 Volume II provides examples of actions underway in communities across the United States to reduce the risks associated with cli­mate change, increase resilience, and improve livelihoods.

The report concludes that Earth’s climate is now changing fast­er than at any point in the history of modern civilization, primarily as a result of human activities. The impacts of global climate change are already being felt in the United States and are projected to intensify in the future—but the severity of future impacts will de­pend largely on actions taken to reduce greenhouse gas emissions and to adapt to the changes that will occur. The following summary information is excerpted from the report.

1.  Key Scientific Advances

The following scientific advances have been made since the previous assessment (NCA3) in 2014.

Detection and Attribution: Significant advances have been made in the attribution of the human influence for individual climate and weather extreme events.

Extreme Events and Atmospheric Circulation: How climate change may affect specific types of extreme events in the United States and the extent to which atmospheric circula­tion in the midlatitudes is changing or is projected to change, possibly in ways not captured by current climate models, are important areas of research where scientific understanding has advanced.

Localized Information: As computing resources have grown, projections of future climate from global models are now being conducted at finer scales (with resolution on the order of 15 miles), providing more realistic characterization of intense weather systems, including hurricanes. For the first time in the NCA process, sea level rise projections incorporate geographic variation based on factors such as local land subsidence, ocean currents, and changes in Earth’s gravitational field.

Ocean and Coastal Waters: Ocean acidification, warming, and oxygen loss are all increas­ing, and scientific understanding of the severity of their impacts is growing. Both oxygen loss and acidification may be magnified in some U.S. coastal waters relative to the global average, raising the risk of serious ecological and economic consequences.

Rapid Changes for Ice on Earth: New observations from many different sources confirm that ice loss across the globe is continuing and, in many cases, accelerating. Since NCA3, Antarctica and Greenland have continued to lose ice mass, with mounting evidence that mass loss is accelerating. Observations continue to show declines in the volume of mountain glaciers around the world. Annual September minimum sea ice extent in the Arctic Ocean has decreased at a rate of 11%–16% per decade since the early 1980s, with accelerating ice loss since 2000. The annual sea ice extent minimum for 2016 was the second lowest on record; the sea ice minimums in 2014 and 2015 were also among the lowest on record.

Potential Surprises: Both large-scale shifts in the climate system (sometimes called “tip­ping points”) and compound extremes have the potential to generate outcomes that are difficult to anticipate and may have high consequences. The more the climate changes, the greater the potential for these surprises.

2. Extreme Events

Climate change is altering the characteristics of many extreme weather and climate-related events. Some extreme events have already become more frequent, intense, widespread, or of longer duration, and many are expected to continue to increase or worsen, presenting substantial challenges for built, agricultural, and natural systems. Some storm types such as hurricanes, tornadoes, and winter storms are also exhibiting changes that have been linked to climate change, although the current state of the science does not yet permit detailed understanding. Individual extreme weather and climate-related events—even those that have not been clearly attributed to climate change by scientific analyses—reveal risks to society and vulnerabilities that mirror those we expect in a warmer world. Non-climate stressors (such as land-use changes and shifting demograph­ics) can also amplify the damages associated with extreme events. The National Oceanic and Atmospheric Administration estimates that the United States has experienced 44 billion-dollar weather and climate disasters since 2015 (through April 6, 2018), incurring costs of nearly $400 billion (

3. Reducing Risks through Adaptation Actions

Key Message 1: Adaptation Implementation Is Increasing Adaptation planning and implementation activities are occurring across the United States in the public, private, and nonprofit sectors. Since the Third National Climate Assessment, implementation has increased but is not yet commonplace.

Key Message 2: Climate Change Outpaces Adaptation Planning Successful adaptation has been hindered by the assumption that climate conditions are and will be similar to those in the past. Incorporating information on current and future climate conditions into design guidelines, standards, policies, and practices would reduce risk and adverse impacts.

Key Message 3: Adaptation Entails Iterative Risk Management Adaptation entails a continuing risk management process; it does not have an end point. With this approach, individuals and organizations of all types assess risks and vulnerabilities from climate and other drivers of change (such as economic, environmental, and societal), take actions to reduce those risks, and learn over time.

Key Message 4: Benefits of Proactive Adaptation Exceed Costs Proactive adaptation initiatives—including changes to policies, business operations, capital investments, and other steps—yield benefits in excess of their costs in the near term, as well as over the long term. Evaluating adaptation strategies involves consideration of equity, justice, cultural heritage, the environment, health, and national security.

Key Message 5: New Approaches Can Further Reduce Risk Integrating climate considerations into existing organizational and sectoral policies and practices provides adaptation benefits. Further reduction of the risks from climate change can be achieved by new approaches that create conditions for altering regulatory and policy environments, cultural and community resources, economic and financial systems, technology applications, and ecosystems.


USGCRP, 2017: Climate Science Special Report: Fourth National Climate Assessment, Volume I. Wuebbles, D.J., D.W. Fahey, K.A. Hibbard, D.J. Dokken, B.C. Stewart, and T.K. Maycock, Eds. U.S. Global Change Research Program, Washington, DC, USA, 470 pp.

USGCRP, 2018: Impacts, Risks, and Adaptation in the United States: Fourth National Climate Assessment, Volume II: Report-in-Brief [Reidmiller, D.R., C.W. Avery, D.R. Easterling, K.E. Kunkel, K.L.M. Lewis, T.K. Maycock, and B.C. Stewart (eds.)]. U.S. Global Change Research Program, Washington, DC, USA, 186 pp.

Quarterly Newsletter November 2018

Climate science, economics and politics

Thomas Mortlock and Ryan Crompton

The science

Twenty-eight years on from the First Assessment Report in 1990, the IPCC’s most recent Special Report on Global Warming delivers an urgent warning to policymakers that we are reaching the point of no return for mitigating anthropogenic impacts on global warming and associated climate change. The report has divided opinion in Australia and further highlights the discord between climate science, economics and politics nationally.

The report finds that limiting global warming to 1.5°C would now require rapid and unprecedented change in all aspects of society. The report highlights that we are already seeing the consequences of 1°C of global warming through more extreme weather, rising sea levels and diminishing Arctic sea ice.

While 1.5°C may seem a small increase, it is how changes in mean temperatures are related to extreme weather events that is important.

A small increase in the mean temperature also shifts the tails of the distribution, meaning it may impact extreme weather events that are temperature-dependent. However, any relationship between increasing mean global air temperature and extreme weather will be complex, and both peril and location-dependent.

Prof Andy Pitman, from the University of New South Wales and Director of the ARC Centre of Excellence for Climate Extremes, described this in an anecdote to BBC News back in January:

“the probability works a bit like if you stand at sea level and throw a ball in the air, and then gradually make your way up a mountain and throw the ball in the air again. The chances of the ball going higher increases dramatically. That’s what we’re doing with temperature.”

As an industry partner of the ARC Centre of Excellence, Risk Frontiers are looking to couple their cat modelling infrastructure to downscaled climate change projections to model the impacts on a peril-by-peril basis for business.

The economics

A day after the IPCC report was published, the Royal Swedish Academy of Sciences awarded William Nordhaus and Paul Romer the Nobel Memorial prize for economics for their work on climate change and economic growth.

Since the 1970s, Prof. Nordhaus has been warning governments that their economic models were not properly taking into account the impact of global warming. Similarly, Prof. Romer developed the “endogenous growth theory”: the notion that countries can improve their underlying performance if they concentrate on supply-side measures such as research and development, innovation and skills. He argues that the creation and spread of ideas – whether that be around climate change or otherwise – is necessary for economic growth.

Understanding the economic costs of climate-related damages is essential to answering the question of how much society should be willing to pay to avert that damage (The Economist, 2018). Prof. Nordhaus’ work addresses this issue by modelling the economic harm of carbon emissions, thus allowing him to estimate the likely economic costs of the different IPCC Representative Concentration Pathways (RCPs).

Prof. Romer believes it is perfectly possible for global warming to be kept to a maximum of 1.5°C:

“Once we start to try to reduce carbon emissions, we’ll be surprised that it wasn’t as hard as we anticipated. The danger with very alarming forecasts is that it will make people feel apathetic and hopeless. One problem today is that people think protecting the environment will be so costly and so hard that they want to ignore the problem and pretend it doesn’t exist.”

The politics

The IPCC report is published at a time of international discord on climate mitigation, with most scientists acknowledging that the likelihood of achieving a plateau at the proposed 1.5 °C is very small. This is essentially a reflection on the myopic nature of global political institutions, and the opposing long-term nature of the problem at hand.

It also highlights the divisive nature of climate change in Australia. As elsewhere, it has become entangled with political agendas, class, energy and living standards. However, unlike elsewhere, adaptation to climate change has yet to occupy an ongoing, cross-party role in government policy as it has done, for example, in Europe. It has exposed an interesting divide between sectors that has come to the fore in recent years – with banking, insurance and industry at large leading the charge in understanding climate change risk and exposures. APRA, the Australian Prudential Regulation Authority, has ensured momentum here when in February 2017 Geoff Summerhayes declared climate change can no longer be considered a future financial risk:

“While climate risks have been broadly recognised, they have often been seen as a future problem or a non-financial problem. The key point I want to make today, and that APRA wants to be explicit about, is that this is no longer the case. Some climate risks are distinctly ‘financial’ in nature. Many of these risks are foreseeable, material and actionable now. Climate risks also have potential system-wide implications that APRA and other regulators here and abroad are paying much closer attention to.”

The bottom line

The global impasse on mitigation efforts only serves to highlight the importance of climate change adaptation planning and risk management in Australia. We’ll need to continue to adjust to the effects of climate change in the absence of addressing the underlying sources, but, ideally, we’d do both. If we believe the economics of the two most recent Nobel prize recipients, this may not be as costly as we think.


APRA (2017). Australia’s new horizon: climate change challenges and prudential risk. Geoff Summerhayes, Executive Member – Insurance Council of Australia Annual Forum, Sydney.

BBC News (2018). How Australia’s extreme heat might be here to stay. BBC News article by Adam Morton, Hobart, 13 January 2018.

IPCC (2018). The special report on global warming of 1.5°C: summary for policymakers.

The Economist (2018). The Nobel prize for economics is awarded for work on the climate and economic growth. Economist Business and Finance article, 8 October 2018.

Responses to the Lombok Earthquake, 2018 – Rapid Assessment Study

Jonathan van Leeuwen, Andrew Gissing and Ashley Avci

The recent earthquake that occurred in Lombok in August, 2018, presented an opportunity to study the responses of those affected in the immediate aftermath of the event. We find that tourists caught up in disasters are uniquely vulnerable. Few followed the encouraged actions of what to do in the event of an earthquake and most were reliant on local residents and tourist operators for advice. This article summarises the earthquake and how people responded and provides some reflections for policy makers.


The island of Lombok is located in the Nusa Tenggara Barat Region of Indonesia. It lies on the boundary between the Australian Plate and the Sunda Plate, which has produced numerous powerful earthquakes in the past. The region is a popular tourist destination, with rapidly increasing numbers of visitors since developed countries lifted travel warnings following the 2002 and 2005 Bali bombings and the SARS (severe acute respiratory syndrome) outbreak between 2002 and 2004. Tourism is a major source of income for Lombok and the neighbouring Bali and Gili Islands, with millions of visitors from around the world each year.

The August 5, 2018, Mw6.9 earthquake occurred as the result of shallow thrust faulting on or near the Flores Back Arc Thrust. The earthquake occurred in a subduction plate boundary region where the Sunda and Australia plates converge (USGS, 2018). In the region surrounding the location of the earthquake, there have been six other events of Mw6.5 or larger over the previous century. Four of these are likely to have occurred on the Back Arc Thrust system: a Mw6.5 in the Bali region to the west of Lombok in July 1976 and three events of Mw6.5, Mw6.5 and Mw6.6 in the Sumbawa region to the east of Lombok in November 2007 and November 2009. The Sumbawa earthquakes were associated with several deaths, hundreds of injuries and the destruction of hundreds of houses. This history of recent earthquakes means that locals would have been familiar with the impacts of damaging earthquakes.

Figure 1: Earthquake location with regional context

The earthquake occurred at a depth of 31.0km, centred at the northern tip of Lombok. The local time was 7:46pm. It was preceded by a main foreshock on July 29, 2018 of Mw6.4, and numerous aftershocks including a Mw5.9 event on August 9, 2018 (USGS, 2018). The earthquake caused severe shaking in Lombok and surrounding islands, including Bali and the Gili Islands, and was felt as far as Sumbawa in the east (Cochrane, 2018) and Trenggalek Regency in the west (Solichah, 2018). Following the earthquake, tsunami warnings were issued: however, the maximum expected height was only half a metre and the warning was later cancelled.

Most of those affected by the earthquake were in North Lombok, East Lombok and Mataram City. Reports indicated that there were 392 fatalities, 1,353 injuries and damage to 67,875 houses, 606 schools, six bridges, three hospitals, ten health centres, 15 mosques, 50 prayer rooms and 20 office units (Badan Nasional Penanggulangan Bencana, 2018).

It is important for emergency managers to have an understanding of human behaviour during extreme events so that they can best develop their plans. In an effort to understand the behaviour of tourists and others following the earthquake, researchers from Risk Frontiers conducted a rapid assessment study utilising media analysis containing interviews with survivors. The method involved locating some 120 news articles sourced from a variety of online international, national and local media outlets. From these articles, interviews with 146 people who experienced the earthquake were extracted and analysed to identify damage that occurred and how people behaved during and after the earthquake.


A significant majority of interviewees were tourists (n=102), who conducted interviews with media outlets from their home countries either remotely or after returning home. Other interviews included local residents (n=20) and expats (n=9), with a further ten not stating where they were from.

At the time of the earthquake, interviewees were located on the island of Bali, approximately 50km to the east of Lombok (n=54); on Lombok (n=40); on the Gili Islands (n=28) and at other locations in the area (n=3). Twenty of those interviewed did not state where they had been at the time of the earthquake.

The interviewees came from a variety of nations, including Australia (n=38), Indonesia (n=25), Britain (n=23), Ireland (n=9), New Zealand (n=8), America (n=7), Singapore (n=5), France (n=4), South Africa (n=3), Canada (n=2), and one interviewee each from Africa (country unstated), Belgium, Denmark, the Netherlands, Malta, Pakistan and Spain. The age of interviewees was captured either by statement in the article or by approximation if a photo was available. Of those interviewed, 99 were categorised between 18 and 60 years old. Three were recorded as above 60 years old, and one was less than 18.

Most interviewees said they were with other adult/s when the earthquake occurred (n=51), or with both children and adult/s (n=24). Ten interviewees said they were with someone but didn’t specify their age/s and nine said they were alone.

In relation to their location at the time of the earthquake, interviewees stated that they were at a restaurant (n=29), in a hotel (size not stated) (n=14), at home and awake (n=9), in a single storey hotel (n=8), in a multi storey hotel (n=7), at home and asleep (n=3), in a shop or shopping centre (n=3), at the beach (n=2), on a footpath (n=2), in a car (n=1) or on a boat (n=1).

Consequences the interviewees observed from the earthquake included collapsed buildings (n=45), debris/objects falling (n=25), injuries (n=23), power cuts (n=22), loss of water from swimming pools (n=15), cracked walls (n=13), food shortages (n=11), deaths (n=11), water shortages (n=6), broken glass (n=4), downed cables (n=4), loss of sanitation (n=2), ground subsidence or uplift (n=1), flooding (n=1) and fires (n=1). Those located in Lombok and the Gili islands observed the most significant damage.

During the earthquake, interviewees most commonly reported, of their own behaviour, that they ran outside (n=43). Others reported that they dropped to the ground as they could not remain standing (n=5), sheltered under a table or bed (n=5), ran outside onto the beach (n=4), moved away from buildings (n=4), sheltered in doorways (n=3), deliberately dropped to the ground (n=2) or moved away from trees (n=2).

During the earthquake, interviewees observed others most commonly either running from buildings (n=44) or screaming (n=37). Other observed behaviours were crying (n=10), moving away from buildings (n=7), caring for others (n=4), running specifically to the beach (n=4), seeking shelter under tables or beds (n=3), holding onto objects or other people (n=3), panicking (n=3), seeking shelter under doorways (n=2), calling or messaging others (n=2) and dropping to the ground (n=1), reporting they could not stand.

Immediately after the earthquake, those interviewed moved to higher ground (fearing a tsunami) (n=29), sought advice on what to do from locals (n=9) or from hotel reception/staff (n=6), gave first aid to the injured (n=4), called or messaged someone (n=4), informed others of tsunami threat levels (n=3), climbed trees (fearing a tsunami) (n=3), searched for family member/s or friend/s (n=3), put on life jackets (fearing a tsunami) (n=2), assisted rescuing trapped person/s (n=2), were themselves incapacitated/requiring treatment (n=2) or extinguished fires (n=1).

Interviewees observed that immediately after the earthquake, others moved to higher ground (n=24), were screaming (n=12), panicking (n=11), caring for others (n=10), running (n=5), assisting the injured (n=5), crying (n=5), remaining on the beach (n=4), calling others (n=4), climbing trees (n=3), searching for others (n=3) or moving debris (n=1).

People said their actions immediately after the earthquake were directed by local residents (n=9), hotel staff (n=9), local authorities (n=3), other tourists (n=2) or by a minister of religion (n=1).

For those interviewees who said they contacted someone, contacts included their parent/s (n=8), other relative/s (n=3), friend/s (n=2), spouse/partner (n=2), children (n=1), authorities (n=1), neighbour/s (n=1) and a stranger (n=1). Six people contacted someone but did not specify who.

Many of those interviewed were not from countries which are associated with high earthquake risk. People’s previous experiences of earthquakes or education provided in their country of origin may have influenced some responses. This possibility is evidenced by the following responses:

“Everyone I spoke to just wants to get out but there’s not one free seat out of here today. About 90 per cent of us were westerners and we’re not trained for how to react in this situation.” (Interviewee from a country not prone to quakes) (Darvall and O’Shea, 2018).

“It’s scary when the ground is buckling under your feet. My partner and I were out of bed and under the table in a flash and we then immediately evacuated the house. When I was a child at school we had earthquake drills. Best training ever.” (Interviewee from a quake-prone country) (NZ Herald, 2018).

Descriptions of interviewees’ emotions during and immediately after the quake included feeling fearful (n=37), panicked (n=14), calm (n=9), concerned (n=7), upset (n=5), terrified (n=5), in shock (n=4), apathetic (n=2), surreal (n=2), and other (n=6). Some 83 interviewees did not state their emotions during and immediately after the quake.

Interviewees said they obtained information about tsunami risk from local residents (n=9), the internet (n=6), warning sirens (or the lack thereof) (n=4), social media (n=3), hotel staff (n=2), calling family or friends at home (n=2), other tourists (n=2), local authorities (n=2), observing the ocean (n=2) or overhearing other people (n=2).

Over subsequent days, a significant number of people said they evacuated soon after (n=29). Some stayed to assist rescue, medical or relief efforts (n=11), although these were mainly locals and expats.

The evacuation of tourists from the Gili Islands was said to be chaotic due to the combination of the lack of capacity to evacuate tourists and the fearful state of tourists and locals. There were reports of long waits, pushing and shoving and passage being offered to the highest bidders. Those interviewees who experienced the evacuation described it as:

“People were just throwing their suitcases on board and I had to struggle to get my husband on, because he was bleeding.” (Embury-Dennis, 2018).

“We just witnessed one of the boats get completely overfilled with tourists climbing on, with the officials trying to keep them back off the boat, pushing them and shoving them. That boat still hasn’t left yet.” (ABC News, 2018).

“People are punching and hitting each other.” (Osborne, 2018).

Discussion and conclusion

Many tourist destinations both within Australia and abroad are susceptible to a range of natural hazard risks. For example, some 26 Australians lost their lives during the Asian tsunami in 2004.

Often, many of the elements that make locations aesthetically appealing to tourists are associated with natural hazard risk. For example, warm, shallow seas and sandy islands make idyllic tropical resort getaways, but these places are often at risk from severe weather, while scenic mountain vistas are often the product of tectonic activity which causes earthquakes and volcanism.

Tourists are uniquely vulnerable. Tourists may be unaware of risks present at their destination, lack local support networks and encounter cultural and communication barriers. Research has previously shown that tourists behave differently to locals. During evacuations they tend not to shelter with family and friends, but seek shelter at public evacuation centres, simply return home or find another hotel (Drabek, 1999). Observations from the Lombok disaster support such conclusions: in particular, that many tourists simply leave soon after a disaster and are reliant on locals for direction.

Many of those interviewed ran from buildings or observed others running from buildings. This behaviour is in conflict with actions encouraged by international and local authorities, which promote the actions of drop, cover and hold.

Counter to some research that suggests that people do not panic in the aftermath of disasters (Lorenz et al. 2018), observations from this event show that panic and chaos can occur. This suggests that in more extreme and less predictable events, panic and chaos is likely or that tourists are more likely to panic. Such questions require further exploration.

Promotion of disaster risk by travel agents and tourism operators conflicts with wider tourism promotion. The Australian Department of Foreign Affairs does provide some details about natural hazard risk on its Smartraveller website, although more needs to be done than passively informing travellers. There could be an opportunity to engage with the medical profession and travel health clinics to promote natural hazard risk and safety behaviours at the time travellers seek travel health advice.

Finally, tourists in Australia are not immune from the impacts of natural hazards, as illustrated by the impacts of Cyclone Debbie in the Whitsunday Region. It is important that tourism operators are engaged regarding disaster preparedness and connected with disaster management structures.


ABC News (2018). Witness describes chaos when earthquake hit the Gili Islands. ABC News. 7 August 2018. [Accessed 8 Oct. 2018].

Badan Nasional Penanggulangan Bencana (2018). Korban Gempa Lombok Terus Bertambah, 392 Orang Meninggal Dunia. [online] Available at: [Accessed 31 Aug. 2018].

Cochrane, J. (2018) Powerful Indonesia earthquake kills at least 82. The New York Times. 5 August 2018. [Accessed 10 Oct. 2018].

Darvall, K and O’Shea, M. (2018). ”I just want to get off the island now”: Bondi restaurant owner shares terrifying evacuation footage after earthquake hit Lombok while he was holidaying with his girlfriend. Daily Mail Australia. 6 August 2018. [Accessed 8 Oct 2018].

Drabek, T.E. (1999). Disaster evacuation responses by tourists and other types of transients. International Journal of Public Administrations, 22(5), pp.665-677.

Embury-Dennis, T. (2018). Lombok earthquake: tourists “forced to pay to board rescue ships”. The Independent. 6 August 2018. [Accessed 8 Oct. 2018].

Lorenz, D.F., Schulze, K. and Voss, M., (2018) Emerging citizen responses to disasters in Germany. Disaster myths as an impediment for a collaboration of unaffiliated responders and professional rescue forces. Journal of Contingencies and Crisis Management, 26 (3), pp.358-367.

Osborne, S. (2018). Lombok earthquake latest: tourists flee Indonesian island after powerful magnitude-7 quake kills at least 98. The Independent. 6 August 2018. [Accessed 8 Oct. 2018].

Solichah, Z. (2018). Warga Jember dan Banuwangi rasakan gempa Lombok. Antara News. 5 August 2018. [Accessed 10 Oct. 2018].

USGS (2018). M 6.9 – 0km SW of Loloan, Indonesia. [online] Available at: [Accessed 31 Aug. 2018].

NZ Herald (2018). New Zealanders in Indonesia describe the violence of Lombok earthquake. NZ Herald. 8 August 2018. [Accessed 8 Oct. 2018].

The 2018 Lake Muir Earthquakes

Paul Somerville, Risk Frontiers.

A second earthquake with magnitude larger than 5 occurred today (November 9, 2018) near Lake Muir in southwestern Western Australia, and Geoscience Australia assigned it a magnitude of 5.4.  This earthquake, shown by the large red dot in Figure 1, occurred about 10 km southeast of the magnitude 5.7 earthquake that occurred on 17 September 2018.   The aftershocks of today’s earthquake, shown by small red dots in Figure 1, are located at the southern end of the aftershock zone of the September 17 event, shown by the yellow dots. Today’s earthquake was preceded by a series of foreshocks that occurred overnight, and was felt between Albany and Perth. The shaking intensities of the two earthquakes are shown in Figures 2 and 3.

Figure 1. Locations of earthquakes near Lake Muir. Source: Geoscience Australia
Figure 2. Shakemap of the 9 November Lake Muir earthquake. Source: USGS
Figure 3. Shakemap of the 17 September Lake Muir earthquake. Source: USGS

The orientation of the fault plane on which the 17 September earthquake occurred is shown on a Wulf net projection in Figure 4. This shows that the earthquake had a thrust mechanism on a fault plane striking east-northeast. The InSAR (interferometric Synthetic Aperture Radar) map in Figure 5 shows that the west side moved up and the east side moved down on a plane dipping down to the west-northwest.  The change in elevation along this dip direction is shown at the upper left of Figure 5.

Figure 4. Focal mechanism of the 17 September 2018 earthquake. Source: USGS
Figure 5. Uplift on the west side and subsidence on the east side of the fault that caused the 17 Sept 2018 earthquake. The uplift and subsidence profile along the line A-B as shown in the inset at top left.

The two earthquakes are shown by the yellow stars on a map of historical earthquake epicentres in the Southwest Western Australia Seismic Zone (SWWASZ) in Figure 6.  The contours show annual probability of events of magnitude 5 and above from QuakeAUS6. The two earthquakes occurred on the southwestern edge of the SWWASZ.

Figure 6. Locations of the Lake Muir earthquakes (yellow stars) on the southwestern edge of the Southwest Western Australia Seismic Zone. Source: QuakeAUS6

As described in our Briefing Note 373,  we have recently released our new probabilistic earthquake loss model for Australia, QuakeAUS 6.0. The updated model, developed by Dr Valentina Koschatzky with input from Risk Frontiers’ Chief Geoscientist, Dr Paul Somerville, incorporates Geoscience Australia’s recent revision of the Australian Earthquake Catalogue (Allen et al., 2017), which has more than halved the rate of earthquakes exceeding 4.5 in magnitude.  The main features of the new model are:

  • New Distributed Earthquake Source Model (based on RF analysis of the new GA catalogue – 2018)
  • Inclusion of an Active Fault Model
  • Updated Soil Classification (McPherson 2017)
  • Updated Soil Amplification Model (Campbell & Bozorgnia 2014)
  • Updated Variable Resolution Grid (VRG), Exposure & Market Portfolio
    (Gnaf_201802 + Nexis 9)

Mystery of the cargo ships that sink when their cargo suddenly liquefies

After the Christchurch earthquake sequence we are very aware of liquefaction and the large scale damage it was responsible for. It may come as a surprise to learn that liquefaction is a big safety issue for the shipping industry where it is sometimes called dry cargo liquefaction or dynamic separation. The article below was written by Susan Gourvenec a professor of offshore geotechnical engineering at the University of Southampton. It was printed in Ars Technica, a website covering news and opinions in technology, science, politics, and society. It publishes news, reviews, and guides on issues such as computer hardware and software, science, technology policy, and video games.

A lot is known about the physics of the liquefaction, yet it’s still causing ships to sink.

Boats carrying grain on the Great Lakes in November 1918

Think of a dangerous cargo, and toxic waste or explosives might come to mind. But granular cargoes such as crushed ore and mineral sands are responsible for the loss of numerous ships every year. On average, 10 “solid bulk cargo” carriers have been lost at sea each year for the last decade.

Solid bulk cargoes – defined as granular materials loaded directly into a ship’s hold – can suddenly turn from a solid state into a liquid state, a process known as liquefaction. And this can be disastrous for any ship carrying them – and its crew.

In 2015, the 56,000-tonne bulk carrier Bulk Jupiter rapidly sank around 300km south-west of Vietnam, with only one of its 19 crew surviving. This prompted warnings from the International Maritime Organization about the possible liquefaction of the relatively new solid bulk cargo bauxite (an aluminum ore).

A lot is known about the physics of the liquefaction of granular materials from geotechnical and earthquake engineering. The vigorous shaking of the Earth causes pressure in the ground water to increase to such a level that the soil “liquefies.” Yet despite our understanding of this phenomenon (and the guidelines in place to prevent it occurring), it is still causing ships to sink and take their crew with them.

Solid bulk cargoes

Solid bulk cargoes are typically “two-phase” materials, as they contain water between the solid particles. When the particles can touch, the friction between them makes the material act like a solid (even though there is liquid present). But when the water pressure rises, these inter-particle forces reduce and the strength of the material decreases. When the friction is reduced to zero, the material acts like a liquid (even though the solid particles are still present).

A solid bulk cargo that is apparently stable on the quayside can liquefy because pressures in the water between the particles build up as it is loaded onto the ship. This is especially likely if, as is common practice, the cargo is loaded with a conveyor belt from the quayside into the hold, which can involve a fall of significant height. The vibration and motion of the ship from the engine and the sea during the voyage can also increase the water pressure and lead to liquefaction of the cargo.

When a solid bulk cargo liquefies, it can shift or slosh inside a ship’s hold, making the vessel less stable. A liquefied cargo can shift completely to one side of the hold. If it regains its strength and reverts to a solid state, the cargo will remain in the shifted position and cause the ship to permanently tilt (or “list”) in the water. The cargo can then liquefy again and shift further, increasing the angle of list.

At some point, the angle of list becomes so great that water enters the hull through the hatch covers, or the vessel is no longer stable enough to recover from the rolling motion caused by the waves. Water can also move from within the cargo to its surface as a result of liquefaction and subsequent sloshing of this free water can further impact the vessel’s stability. Unless the sloshing can be stopped, the ship is in danger of sinking.

The International Maritime Organization has codes governing how much moisture is allowed in solid bulk cargo in order to prevent liquefaction. So why does it still happen?

The technical answer is that the existing guidance on stowing and shipping solid bulk cargoes is too simplistic. Liquefaction potential depends not just on how much moisture is in a bulk cargo but also other material characteristics, such as the particle size distribution, the ratio of the volume of solid particles to water and the relative density of the cargo, as well as the method of loading and the motions of the vessel during the voyage.

The production and transport of new materials, such as bauxite, and increased processing of traditional ores before they are transported, means more cargo is being carried whose material behavior is not well understood. This increases the risk of cargo liquefaction.

Commercial agendas also play a role. For example, pressure to load vessels quickly leads to more hard loading even though it risks raising the water pressure in the cargoes. And pressure to deliver the same tonnage of cargo as was loaded may discourage the crew of the vessel draining cargoes during the voyage.

What’s the solution?

To tackle these problems, the shipping industry needs to better understand the material behavior of solid bulk cargoes now being transported and prescribe appropriate testing. New technology could help. Sensors in a ship’s hold could monitor the water pressure of the bulk cargo. Or the surface of the cargo could be monitored, for example using lasers, to identify any changes in its position.

The challenge is developing a technology that is cheap enough, quick to install and robust enough to survive loading and unloading of the cargo. If these challenges can be overcome, combining data on the water pressure and movement of the cargo with information on the weather and the ship’s movements could produce a real-time warning of whether the cargo is about to liquefy.

The crew could then act to prevent the water pressure in the cargo rising too much, for example, by draining water from the cargo holds (to reduce water pressure) or changing the vessel’s course to avoid particularly bad weather (to reduce ship motions). Or if those options are not possible, the crew could evacuate the vessel. In this way, this phenomenon of solid bulk cargo liquefaction could be overcome, and fewer ships and crew would be lost at sea.

Susan Gourvenec is a professor of offshore geotechnical engineering at the University of Southampton. This article was originally published on The Conversation and has been lightly edited to conform to Ars Technica style guidelines. Read the original article.