Flood Deaths in the Northern Territory

Alice Carney1, Lucinda Coates1,2,3 and Katharine Haynes1,3

1 Macquarie University
2 Risk Frontiers
3 Bushfire and Natural Hazards Cooperative Research Centre

Risk Frontiers recently examined the circumstances surrounding deaths from flood events in Australia as part of a wider Bushfire and Natural Hazards CRC (BNHCRC)-funded project, An analysis of human fatalities and building losses from natural disasters. One of the results found was a heightened level of risk in the Northern Territory. We decided to investigate this a little more closely.


In the previous Risk Frontiers research project, 1859 individually identified flood-related deaths were recorded in Australia from 1900 to 2015 and, of these, 79% were males (Haynes et al., 2016). Death rates showed a steep statistically significant decline up to 1960, with a lesser, steadier decline over the most recent 55 years (Haynes et al., 2016).

Queensland and New South Wales accounted for 75% of the total fatalities across Australia (Haynes et al., 2016). However, when deaths were examined in relation to population size, a heightened level of risk in the Northern Territory (NT) was revealed, with a death rate almost double that of the jurisdiction with the next highest fatality rate (Haynes et al., 2016). When fatalities in the various jurisdictions were examined longitudinally, an expected downward trend in deaths over time was observed, apart from in the NT – particularly in more recent years, where an increasing proportion of flood deaths were seen (Haynes et al., 2016).

This warranted further investigation. This briefing note summarises the results obtained when the demographic characteristics of flood-related deaths occurring in the NT from 1960-2015 were examined.

Fatality totals and trends

From 1960 to 2015 there have been at least 27 fatal floods in the NT, claiming 38 lives. Annual flood fatalities are increasing with time, while death rates have remained constant (figure 1. Note: zero deaths 1960-1964). Males accounted for 74% of the fatalities. The numbers of both male and female flood fatalities are increasing and, although Australia’s male:female ratio is decreasing, the gap between male and female flood deaths in the Northern Territory is growing, showing no sign of equity in the near future.

Flood fatalities in the Northern Territory, 1960-2015
Figure 1: Flood fatalities in the Northern Territory, 1960-2015

Males tend to be more at risk in flood events due to their risk-taking behaviour: for example, males are over-represented in attempting to cross floodwaters (67%), and undertaking an activity near (100%) or in (80%) floodwaters. In all these activities, males are aware of the flood and undertake the activity nonetheless. On the other hand, females are less likely to take these risks and are over-represented only in carrying out activities not near usual watercourses such as staying at home (71%). These statistics suggest gender-specific approaches must be developed to address the clear differences in causes of death.

In regards to age, males are over-represented in most age brackets. Most (98%) decedents were aged 0-59 years, the age group most at risk being those aged 30-39 years.

The Daly River Drainage Basin has claimed the most lives, accounting for 34% of flood fatalities in NT (figure 2). There have been five fatal floods there since 1960, three of which were high fatality (≥ 3 deaths) events. The Todd River Drainage Basin is the second most dangerous, accounting for 21% of fatalities, and having also experienced five fatal flood events.

Location of flood fatalities in NT by drainage basin 1960-2015
Figure 2: Location of flood fatalities in NT by drainage basin, 1960-2015

Indigeneity was investigated from 2000 onwards. A clear inequity is presented, as the indigenous account for 65% of fatalities from 2000-2015. Indigenous males account for over half (52%) of all flood fatalities in NT. This is alarmingly high in comparison to the group least at risk – non-indigenous females, who account for only 4% of fatalities. In terms of age, those most at risk are the 0-9 year-old non-indigenous and 10-19 year-old indigenous groups. It is clear that, similar to the case of the male population, indigenous persons are more prone to risk-taking activities: the majority (71%) of those crossing a flooded watercourse and 67% of those engaged in an activity in a flooded watercourse were indigenous. Research suggests indigenous persons are more likely to present risk-taking behaviours due to poor education on the risks (Atkinson, 2012). A key to reducing flood fatalities in NT is, therefore, training in flood-safe behaviours targeted to the indigenous population.

The riskiest “activity prior to death” was found to be crossing flooded watercourses, which accounts for over a third (35%) of fatalities in NT: 67% of these were male. A total of 57% of female decedents were attempting to cross floodwaters. The second most risky activity (21%) was being engaged in an activity near floodwaters: males were over-represented (100%). [The results for those engaged in an activity not near a usual watercourse – e.g, being at home – are skewed due to one large event in 1977, in which five people were drowned at a cattle station.]

The familiarity of the decedent with the death location was investigated. The term “familiar” as used in this research refers to being within 10km of one’s house. Locals accounted for 87% of flood fatalities. In relation to activity prior to death, locals were most likely to be crossing floodwaters (29%), engaged in an activity in or near floodwaters (21%) or at home (21%). An analysis of those decedents who died at home clearly showed that they chose not to evacuate when warnings were received. This suggests that behavioural changes must be made through education of the risks of ignoring flood warnings. [Note: a relatively small dataset means these results should be treated with some caution.]

Some 25% of the decedents from 2000-2015 were intoxicated and, of those, 80% were attempting to cross a flooded river and 20% were engaged in activities in flood waters. 80% were male; 80% were indigenous. [Note: a relatively small dataset means these results should be treated with some caution.]

There are a few take-home messages around mitigation and education strategies for the Northern Territory. Appropriate strategies must be developed targeting, especially, indigenous males. The aim should be to educate on the risks floods present and the measures that should be taken to avoid them, such as not attempting to cross, or engaging in activities in or near, floodwaters. Haynes et al (2016) gives insight into potential mitigation strategies which should be modified to best suit the target population. The three key strategies should be to educate, pose consequences and apply structural interventions.


Risk Frontiers employed Macquarie University climate science PACE student Alice Carney to investigate the circumstances surrounding flood deaths in the Northern Territory (NT). PACE is Macquarie University’s Professional and Community Engagement program, which gives students a chance to explore key economic, social and ethical challenges by seeing at first-hand how contemporary organizations (such as Risk Frontiers) address them, allowing them to develop new knowledge and skills and explore future career opportunities.

The work utilised Risk Frontiers’ database PerilAUS and the National Coronial Information System (NCIS) database of coronial data, sourced from the Department of Justice and Regulation, Victoria: a resource of coronial records across Australia from July 2000 onwards.


Atkinson, J. 2012. Anthropometric correlates of reproductive success, facial configuration, risk taking and sexual behaviors among indigenous and Western populations: the role of hand-grip strength and wrist width. In: GALLUP, G. G. & SVARE, B. (eds.). ProQuest Dissertations Publishing.

Haynes, K., Coates, L., Van Den Honert, R., Gissing, A., Bird, D., Dimer De Oliveira, F., D’Arcy, R., Smith, C. & Radford, D. 2016. Exploring the circumstances surrounding flood fatalities in Australia—1900–2015 and the implications for policy and practice. Environmental Science & Policy, 76, 165-176.

National Coronial Information System. (2017). Home – National Coronial Information System.   [online] Available at: http://www.ncis.org.au/ [Accessed 4 Oct. 2017].

Hawaii False Alarm Hints at Thin Line Between Mishap and Nuclear War

The following article, by Max Fisher, appeared in The Interpreter, New York Times, on January 14, 2018, the day after state emergency officials in Hawaii made a false warning to take shelter from an inbound missile threat.   Three days later, Japan’s public broadcaster accidentally sent news alerts that North Korea had launched a missile and that citizens should take shelter. The Japanese broadcaster, NHK, corrected itself five minutes later and apologized for the error on its evening news, initially blaming the J-Alert system but later conceding it was not to blame.  NHK’s swift rectification of its error stands in contrast to the 38-minute delay by officials in Hawaii on Saturday in cancelling warnings of an incoming ballistic missile threat. Notwithstanding common misperceptions to the contrary, there is no legal means to prevent launch once the President of the U.S. has made an order to launch, which can be done without consultation and whose timing follows “launch on launch (by the enemy)” to pre-empt the destruction of ground-based missiles by the enemy. Today’s issue of The Independent has an interesting take on how the Hawaii alarm may have been interpreted in Pyongyang on the following website:


and a very close call is described by former U.S. Secretary of Defence William Perry on this website:


Nuclear experts are warning, using some of their most urgent language since President Trump took office, that Hawaii’s false alarm [on January 13, 2018], in which state agencies alerted locals to a nonexistent missile attack, underscores a growing risk of unintended nuclear war with North Korea. To understand the connection, which might not be obvious, you need to go back to the tragedy of Korean Air Lines Flight 007.

In 1983, a Korean airliner bound from Anchorage to Seoul, South Korea, strayed into Soviet airspace. Air defense officers, mistaking it for an American spy plane that had been loitering nearby, tried to establish contact. They fired warning shots. When no response came, they shot it down, killing all 269 people on board.

But the graver lesson may be what happened next. Though it was quickly evident that the downing had been a mistake, mutual distrust and the logic of nuclear deterrence — more so than the deaths themselves — set Washington and Moscow heading toward a conflict neither wanted. The story illustrated how imperfect information, aggressive defense postures and minutes-long response times brought both sides hurtling toward possible nuclear war — a set of dynamics that can feel disconcertingly familiar today.

Ronald Reagan had taken office in 1981 pledging to confront the Soviet Union. Though he intended to deter Soviet aggression, Moscow read his threats and condemnations — he had declared its government an “evil empire” that must be brought to an end — as preludes to war. Mr. Trump’s White House has issued its own threats against North Korea, suggesting that it might pursue war to halt the country’s nuclear weapons development.

The 1983 shooting down, on its own, might have passed as a terrible mistake. But the superpowers had only fragmentary understanding of something that had happened on the far fringes of Soviet territory. In an atmosphere of distrust, technical and bureaucratic snafus drove each to suspect the other of deception.  Moscow received contradictory reports as to whether its pilots had shot down an airliner or a spy plane, and Soviet leaders were biased toward trusting their own. So when they declared it a legal interception of an American military incursion, American leaders, who knew this to be false, assumed Soviet leaders were lying. Moscow had downed the airliner deliberately, some concluded, in an act of undeclared war.

At the same time, Washington made a nearly perfect mirror-image set of mistakes — suggesting that such misreadings are not just possible, but dangerously likely.  Mr. Reagan, furious at the loss of life, accused Moscow of deliberately targeting the civilian airliner. He denounced Soviet society itself as rotten and in pursuit of world domination.  In fact, a C.I.A. assessment, included in the president’s daily briefing that morning, had concluded the incident was likely an error. Mr. Reagan appeared to have simply missed it.

But Soviet leaders had never considered this; they assumed Mr. Reagan was lying about their intentions. Some concluded he had somehow lured the Soviet Union into downing the aircraft as cover for a massive pre-emptive attack, which they feared might come at any moment.  Each read the other’s blundering and dissembling as intentional, deepening suspicions among hard-liners that the other side was laying the groundwork for war. And if war was coming, the logic of nuclear deterrence all but required firing first.

Nuclear-armed missiles had recently achieved a level of speed and capability so that one power could completely disarm another in a matter of minutes. This created something called first-strike instability, in which firing first — even if you think you might be firing in error — is the only way to be sure of preventing your own obliteration.  The result was that the United States and the Soviet Union repeatedly went to the brink of war over provocations or even technical misreadings. Often, officials had mere minutes to decide whether to retaliate against seemingly real or impending attacks without being able to fully verify whether an attack was actually underway. In the logic of nuclear deterrence, firing would have been the rational choice.

That dynamic is heightened with North Korea, which is thought to have only a few dozen warheads and so must fire them immediately to prevent their destruction in the event of war.  “Today’s false alarm in Hawaii a reminder of the big risks we continue to run by relying on nuclear deterrence/prompt launch nuclear posture,” Kingston Reif, an analyst with the Arms Control Association, wrote on Twitter, referring to the strategy of firing quickly in a war. “And while deterring/containing North Korea is far preferable to preventive war, it’s not risk free. And it could fail.”

If similar misunderstandings seem implausible today, consider that an initial White House statement called Hawaii’s alert an exercise — though state officials say it was operator error. Consider that 38 minutes elapsed before emergency systems sent a second message announcing the mistake. If even Washington was misreading events, the confusion in Pyongyang must have been far greater. Had the turmoil unfolded during a major crisis or period of heightened threats, North Korean leaders could have misread the Hawaiian warning as cover for an attack, much as the Soviets had done in 1983. American officials have been warning for weeks that they might attack North Korea. Though some analysts consider this a likely bluff, officials in Pyongyang have little room for error.

Vipin Narang, a nuclear scholar at the Massachusetts Institute of Technology, suggested another possible scenario, using shorthand terms to refer to the president and his nuclear command systems, which Mr. Trump has nearby at all times. “POTUS sees alert on his phone about an incoming toward Hawaii, pulls out the biscuit, turns to his military aide with the football and issues a valid and authentic order to launch nuclear weapons at North Korea,” Mr. Narang wrote on Twitter, adding, “Think it can’t happen?”

Unlike in 1983, no one died in Hawaii’s false alarm. But deaths are not necessary for a mistake to lead to war. Just three months after the airliner was shot down, a Soviet early warning system falsely registered a massive American launch. Nuclear war may have only been averted because the Soviet officer in charge, operating purely on a hunch, reported it as an error.

North Korea is far more vulnerable than the Soviet Union was to a nuclear strike, giving its officers an even narrower window to judge events and even greater incentive to fire first. And, unlike the Soviets, who maintained global watch systems and spy networks, North Korea operates in relative blindness. For all the power of nuclear weapons, scholars say their gravest dangers come from the uncertainty they create and the fallibility of human operators, who must read every signal perfectly for mutual deterrence to hold.

In 1983, Washington and Moscow took steps that heightened the uncertainty, darkly hinting at each other’s illegitimacy and threats of massive retaliation, in a contest for nuclear supremacy, and survival. Each was gambling they could go to the brink without human error pushing them over. William J. Perry, a defense secretary under President Bill Clinton, called the false alarm in Hawaii a reminder that “the risk of accidental nuclear war is not hypothetical — accidents have happened in the past, and humans will err again.”

Mr. Reagan concluded the same, writing in his memoirs, “The KAL incident demonstrated how close the world had come to the nuclear precipice and how much we needed nuclear arms control.” Mikhail Gorbachev, who soon after took over the Soviet Union, had the same response, later telling the journalist David Hoffman, “A war could start not because of a political decision, but just because of some technical failure.”  Mr. Gorbachev and Mr. Reagan reduced their country’s stockpiles and repeatedly sought, though never quite reached, an agreement to banish nuclear weapons from the world. But Mr. Trump and North Korea’s leader, Kim Jong-un, remain locked in 1983, issuing provocations and threats of nuclear strikes on push-button alert, gambling that their luck, and ours, will continue to hold.

6 ways you can prepare for the on-coming heatwave

Australian Geographic spoke with heatwave risk management experts to determine what you can do to beat the heat over the next week.

Most of south-east Australia is gearing up for what’s predicted to be a sweltering five day heatwave, according to the Bureau of Meteorology, and fire-fighters are on high alert.

Weather risk management experts Andrew Gissing and Lucinda Coates from Risk Frontiers say that when the heatwave hits, it’s important to avoid complacency and have a well thought out plan.

Read more.


Australia’s ‘deadliest natural hazard’: what’s your heatwave plan?

This article by Andrew Gissing and Lucinda Coates has appeared in today’s The Conversation.

“Heatwaves are Australia’s deadliest natural hazard, but a recent survey has found that many vulnerable people do not have plans to cope with extreme heat.

Working with the Bushfire and Natural Hazards Cooperative Research Centre and the Bureau of Meteorology, my colleagues and I surveyed 250 residents and 60 business managers in Western Sydney and the NSW North Coast.

We found that 45% of those at risk – including the elderly, ill and very young – did not proactively respond to heatwave warnings as they did not think it necessary or did not know what to do.”

Follow the link below to read more:


The heat is on: but we’ve been there before

By Lucinda Coates, Risk Frontiers

Sydney was the hottest city on earth on Sunday 7 January 2018 (and no, I’m not talking about its nightlife) when Penrith, in the outer west, reached 47.30C, pipping its previous record set on 11 February 2017 (News Limited, 2018).

But if you want really hot, then travel back in time to 1939, when the Old Richmond Station set Sydney’s official heat record at 47.80C.  Yes, we’ve had heatwaves before.  Way before.

The table below shows numbers of deaths and death rates per 100,000 population from episodes of extreme heat in Australia by decade between 1844 and 2010, as recorded in Risk Frontiers’ PerilAUS database (after Coates et al. (2013)). PerilAUS is a resource of natural hazard event impacts reaching back to the early days of Australia’s European settlement.  The death rate is the number of deaths per head of population in the country at that time, and was consistently significantly higher between 1890 and 1939 than for any period before or since.

Of all of the entrIes in the Table, the January 1939 event was notable for its longevity and record daily temperature maxima. Victoria and South Australia, as well as country NSW, were affected with Melbourne reaching a high of 45.60C and Adelaide 46.10C.  In NSW, Bourke suffered through 37 consecutive days over 380C.

PerilAUS records show that at least 420 people died in the 1939 event across Australia, most (77%) in NSW. The series of heatwaves were accompanied by strong northerly winds, and followed a very dry six months. This led to the disastrous Black Friday bushfires in Victoria, which killed 71 people.

Most will remember the catastrophic bushfires that destroyed several towns in Victoria in 2009 but not many will remember that these fires also followed two heatwave events across Victoria and SA, where at least 432 people died.

This figure comprises mainly a measure of excess deaths rather than recorded individual deaths. An excess death is a premature death and, in this context, a measure of the number of deaths occurring over and above that expected for that location and time of year.

In 2009, new records of three consecutive days over 430C in Melbourne and eight over 400C in Adelaide were set.  A feature of these heatwaves was the very hot minimum temperatures, with Melbourne’s temperature falling to between 20-250C overnight and Adelaide to just 300C.

A similar death toll resulted from the heatwave that occurred from October 1895 to January 1896 that impacted nearly the entire continent but especially the interior. PerilAUS records 435 deaths, 89% of them within NSW.  Deaths also occurred in SA, WA, Victoria and Queensland. Bourke, in NSW, lost 1.6% of its population to the heat: temperatures of 400C in the shade were already being recorded in October, mid-Spring.

Heatwaves in Australia, including catastrophic ones, are not new. Risk Frontiers first noted the fact that they are Australia’s number one natural hazard killer more than two decades ago (Coates, 1996). For further reading on this important natural hazard, the reader is referred to Coates et al. (2013).


Coates L, 1996, An Overview of Fatalities from Some Natural Hazards, Proceedings, NDR96 Conference on Natural Disaster Reduction, 29 September-2 October 1996, Gold Coast, ed. R L Heathcote, C Cuttler and J Koetz. 49-54 http://search.informit.com.au/documentSummary;dn=547566533577889;res=IELENG

Coates L, Haynes, K, O’Brien, J, McAneney, J and Dimer de Oliveira, F, 2014, Exploring 167 years of vulnerability: An examination of extreme heat events in Australia 1844-2010, Environmental Science & Policy, 42:33-44. http://www.sciencedirect.com/science/article/pii/S1462901114000999

News Limited, 2018 – news.com.au [Daily Telegraph, 8 January 2018, originally published as Sydney: The hottest place on Earth], NSW heatwave: Sydney the hottest place on Earth, http://www.news.com.au/technology/environment/nsw-heatwave-sydney-the-hottest-place-on-earth/news-story/0486ad5df9b5025ac24a4507aa1b8a17, accessed 8/1/2018