On 21 April 2015, the town of Dungog in rural NSW suffered experienced a sudden, apparently unprecedented flash flood that claimed three lives.  Between 28 and 31 August the NSW Deputy State Coroner heard evidence about the event, the ability of the Bureau of Meteorology (BoM) to forecast the event and the response by NSW SES. The coroner’s findings were released on 29 September 2017 (Inquest into the deaths of Robin MacDonald, Colin Webb and Brian Wilson (Dungog Floods)).

In NSW SES, a local controller is appointed by the SES Commissioner for a local government area.  He or she is ‘responsible for the control and co-ordination of the activities of SES units in the relevant local government area’ (State Emergency Service Act 1989 (NSW) s 17).  A unit controller is appointed for a particular unit (State Emergency Service Act 1989 (NSW) s 17A). The unit controller for Dungog unit was also the local controller for the Dungog local council area. During these coroner’s proceedings the local/unit controller was separately represented that is a barrister appeared during the inquest to represent the controller’s interests.  The SES was also represented. It is not known who funded the Controller’s legal representation but I would anticipate that it was the SES/NSW Government.

The event

On 17 April 2015 the BoM forecast the development of an East Coast Low and began the process of issuing severe weather alerts for the region.

On the evening of 20-21 April rain fell across the Myall Creek catchment. The BoMs definition of ‘heavy’ rainfall was rain of, or in excess of, 46.9mm/hour at Dungog Post Office or 43.7mm/hour at a gauge in the Upper Myall Creek.  In the 24 hours to 5am the rain had been consistent, but not ‘heavy’.  Between 5am and 7am on 21 April 116.7mm of rain fell at the Dungog Post Office.  It was during those two hours that flooding occurred in Dungog entering homes, including the homes where three people died.  The coroner said (at [128]):

What occurred between about 05:00hours and 07:00hours on 21 April 2015 was an entirely unpredicted, localised weather anomaly involving an extreme rain event centred over the southern part of the Myall Creek Catchment, including the township of Dungog.

The SES had flood action cards to guide incident controllers. The cards predicted inundation of the town at various river heights and when consideration should be given to evacuation.  Various low lying properties were flagged for evacuation if the Williams River was to reach 7m.  At 5am the river was at 5.57m; at 6am it was 6.93m (still below the evacuation trigger point) and at 7am it was at 8m but at that time the reports were that the flooding was receding (see [133]).  The Williams River didn’t peak until 10.30am when it peaked at 8.68m but this was after the flooding in the town had receded.  The flooding in the town was possibly greater than a predicted 1:1000-year flood event.

What impacted Dungog was a flash flood – that is ‘“flooding which is sudden and often unexpected because it is caused by sudden local or nearby heavy rainfall’ ([136]).  The flash flooding here was caused by the rain with a contribution from the sodden ground and the amount of water in the Williams River and Myall Creek. At the time ([139]) ‘The Myall Creek Catchment was not defined as a “flash flood catchment” for the purposes of the NSW State Flood Sub Plan 2009’ that is the threat to the town was understood to come from flood waters from the rivers and creeks.

The search for answers

Residents affected by the flood were critical of the response of the emergency services (see  Donna Page and Dan Proudman, ‘Questions emerge in wake of Dungog flood crisis‘ Newcastle Herald (Online) 22 May 2015).  That story reports:

… questions are mounting about the role of the SES and whether the organisation tasked with being the ‘‘lead combat agency’’ for flood disasters failed to prepare adequately.

Why were elderly residents in flood-prone areas left sleeping in their beds, unaware of the rising floodwaters until it was too late?

What was behind a group of angry residents having to drive to the Dungog SES unit about 6am on Tuesday to demand a boat be put in the water to rescue people?

And why weren’t other emergency services told of the imminent disaster before the town’s fire crews and police were sent 27kilometres away to Stroud as people drowned in town?

At the start of the inquest ‘The head of the New South Wales State Emergency Service … admitted failings in responding to a deadly Hunter Valley storm’ (Giselle Wakatama, ‘SES apologises for failings during deadly Dungog storm’, ABC News (Online) 28 August 2017) but it is not clear from that report what ‘failings’ he was referring to.  The Coroner noted (at [31]-[32]) that:

As at 20 April 2015 [the Unit/Local Controller] had not received:

i. a formal handover briefing;
ii. training in the role of Local Controller and/or the Unit Controller;
iii. training in relation to the Dungog Shire Local Flood Plan 2011;
iv. any specific training in the Dungog SES Unit Flood Acti-Cards regarding responses to flooding in the Paterson and Williams Rivers; or
v. any AIIMS training, including training to perform the AIIMS Incident Management Role of “Incident Controller”.

As at 20 April 2015 the SES did not offer:

i. a system for conducting a handover briefing for in-coming and outgoing Local Controllers and/or Unit Controllers; or
ii. routine provision of AIIMS training, in particular for the role of the AIIMS Incident Management Role of “Incident Controller”.

Further, the Deputy Local Controller (who was for part of the relevant time acting as the incident controller):

… had not been provided with training as the SES Dungog Deputy Local Controller and/or Deputy Unit Controller; or with AIIMS training, including training in relation to the performance of the AIIMS Incident Management Role of “Incident Controller” ([37]).

The Coroner did not address those issues beyond identifying them.  She did not suggest that any of those failings contributed to the deaths.

The article in the Newcastle Herald said:

But flooding in Dungog was in no way a novelty and authorities were aware of the risks from Myall Creek.

The SES has been responsible for floods and storms since being created following the 1955 Maitland floods.

The combat agency’s own Dungog Shire Local Flood Plan, obtained by the Herald, identifies nine houses, two businesses and four aged-care units as high risk of flooding that may require evacuation.

That was true, but on the Coroners finding the flooding was so fast that it had happened before the river gauges indicated that evacuation was warranted.

Other issues raised by the Newcastle Herald involving inter-agency communication, whether emergency services were sent to another town when they were required in Dungog and inter-agency cooperation were not addressed by the Coroner.


In the Newcastle Herald, one person is quoted as saying ‘‘‘The residents had each other and that was it,’’ she said.  ‘‘We certainly didn’t get any help.’’’  Another

… said the SES was absent throughout most of the crisis and terribly slow off the mark when the situation exploded.

He said for hours residents had to fend for themselves because there was no help coming.

It is axiomatic that first responders are always local – and it was members of the community who stepped up to help their neighbours before any emergency services could be called or could attend.  The coroner reported on many of those rescues including one man who helped at least 4 people to safety and recovered the body of one of the deceased ([106]-[127]).  In her conclusions the Coroner paid tribute to the rescuers. She said (at [172]):

In closing, I would like to acknowledge the following:

i. the courageous efforts of the residents of Dungog. Were it not for the conduct of a number of people, it is likely that more lives would have been lost;

She also acknowledged:

ii. the involvement of emergency services officers involved in the flood response across the Hunter region, including from NSW Police, Fire and Rescue NSW and the SES;…

The BoM

With respect to the BoM the Coroner said that she was:

… satisfied on all of the evidence … that the extent of rain over the township of Dungog and the catchment of the Myall Creek, and the resultant flooding, was an unpredictable and unpredicted weather anomaly.


The Coroner said (at [158]-[159]):

… the significant flood effects which occurred on the morning of 21 April 2015, all occurred, effectively without warning, in the period between about 6:16am and about 7.00 am that morning.

In the circumstances, on all of the evidence, there is no basis to conclude that either [the local/unit controller or deputy local/unit controller] should have given an evacuation warning or issued an evacuation order any time before about 6.16am on 21 April 2015.

In other words, there was nothing the BoM or the SES could have, or should have done to avoid the deaths in this tragic event.

The future

With respect to avoiding future deaths the Coroner noted:

  • The flooding risk for Dungog is now better understood. ‘This flood intelligence is able to inform flood planning for Dungog going forward’ ([140]-[146]);
  • At the time of the event the BoM maintained automatic rain fall gauges that gave near real time information. River heights however were monitored by gauges maintained by the NSW Office of Water (NOW). Information on river heights was fed to the NOW every hour and then to the BoM with the result that the river height information received by the BoM, and used to guide flood watches and flood warnings, could be 1 to 1 ½ hours delayed.  The BoM and the National Flood Infrastructure Working Group were working on improving river height monitoring to give real time data ([161]).
  • Dungog Council had purchased five low lying properties, including four where homes had been washed away, to take those properties out of the housing stock ([163])
  • The Council has also revised its Flood Planning Area and supported the demolition of six aged care units that are now in the Flood Planning Area ([163]);
  • The SES had improved hand-over and training for local and unit controllers and recruited new members for the Dungog unit ([164]).

The Coroner did identify that the evidence showed ‘the need for an automated flash flood/flood early warning system for Dungog’ ([165]).  The Coroner also noted that the SES had worked with a meteorologist during an East Coast Low in 2016 and that having a meteorologist on hand was ‘extremely beneficial’.  Because of these findings, the Coroner made two recommendations.  They were:

That the NSW State Emergency Service, the Bureau of Meteorology and the Dungog Shire Council work together to convene a technical advisory group involving representatives from each organisation, and liaise with any officer of the Office of Environment and Heritage, and any consulting engineer(s) and local flood expert(s) engaged from time to time, to look at solutions for warning and responding to flood and flash flood events in Dungog (including the Myall Creek catchment):

(i) On an interim basis while an automated flood warning system is developed; and

(ii) On a long term basis, to consider developing an automated flood warning system designed to use a combination of rainfall and riverine water levels relevant to flood in the Myall Creek and its tributaries.


That further consideration be given to providing the NSW State Emergency Service with access to an out-posted meteorologist from the Bureau of Meteorology for ongoing planning and consultation, on a part-time basis, as well as assistance during weather events.


This was a tragic event but the evidence before the Coroner was that the BoM and the NSW SES had conducted extensive post event studies to understand the processes that led to the flooding. Given the information that they had and the significant, localised nature of the rain event the Coroner was satisfied that there was nothing more either the BoM or the SES could or should have done to prevent the loss of life.