AN INQUIRY INTO THE FIRE AT THE SYDNEY BUILDING

CORONERS COURT OF THE AUSTRALIAN CAPITAL TERRITORY Case Title: AN INQUIRY INTO THE FIRE AT

THE SYDNEY BUILDING Citation: [2018] ACTCD 15 Date of Findings: 1 November 2018 Before: Coroner P. J. Morrison Decision: 1. A fire occurred at Sydney Building, London Circuit, Canberra City on 17 February 2014.

2. The point of origin of the fire was that part of the building occupied by the business known as Izakaya Coo 2 G 4 U and was caused by an unattended cooking wok on a gas stove which in turn ignited the cooking oil within the wok.

3. The cause and origin of this fire are sufficiently disclosed and a hearing is unnecessary.

4. Pursuant to s 52(4)(a)(i) of the Coroners Act 1997, two matters of public safety are found to arise in connection with this inquest.

Legislation Cited: Coroners Act 1997 (ACT) s 52(4)(a)(i) File Number: CF 178 of 2014

1. On 14 July 2016 I made interim findings as to the cause and origin of this fire: A. A fire occurred at Sydney Building, London Circuit, Canberra City on 17 February 2014.

B. The point of origin of the fire was that part of the building occupied by the business known as Izakaya Coo 2 G 4 U and was caused by an unattended cooking wok on a gas stove which in turn ignited the cooking oil within the wok.

2. Completion of the Inquiry at that time was pending further investigations into a possible matter of public safety as per paragraph 17 of these reasons. This process has been long and complex, but I believe necessary to fulfil my statutory functions.

3. Pursuant to s 52(4)(a)(i) of the Coroners Act 1997, I now confirm a matter of public safety is found to arise in connection with this Inquiry.

Circumstances of the Fire

4. I find the circumstances of the fire to have been as follows.

5. On 17 February 2014 Eric Hwang commenced work at Izakaya Coo 2 G 4 U (“Coo”) at about 7:30am. After doing some food preparation work, Mr Hwang started the deep fryers in the kitchen (probably a little before 8am). These are a deep fryer and a metal wok filled with oil. Mr Hwang turned the pilot on the deep fryer on and turned on the gas under the wok so that there was only a very little flame to warm up in preparation for frying. Mr Hwang left the restaurant twice that morning to go to the IGA to buy things. Mr Hwang later advised police that ordinarily when he leaves the restaurant he normally turns the gas off so the flame goes out and he just leaves the pilot for the stove on.

6. At about 9:00am another staff member, Tae, arrived at Coo. A short time after that, possibly around 9:15am, Tae drove Mr Hwang to Dickson so that Mr Hwang could attend to some personal business. When asked by police subsequently, Mr Hwang could not remember whether he turned the gas burner to the wok off before leaving with Tae, but he advised police that he was always very careful and always double checked to ensure the burners are off every time before leaving the restaurant.

7. However, investigation of the site after the fire located evidence that one of the gas fired wok tables (next to the deep fryer, with a wok sitting on top of it) was left on. On first sighting the handle on the appliance was found in the off position. However, when the valve was pulled apart, the rubber seal in the gas control valve for the wok table had melted around the ball valve at a point which corresponded to the valve being open at about 1/3 from fully open and the position of the handle being at about 45 degrees from off. Additionally, the sit valve (before the gas ball valve) was found in the on position. The fact of the handle on the appliance being found in the off position is explicable by virtue of the safety feature on the valve whereby in the presence of heat the weight of the handle is sufficient to close the valve to make it safe.]

8. The first call (via 000) made to ACT Fire & Rescue (“ACTF&R”) reporting a fire is at 9:44 am, but multiple calls were received. The first fire officers were in attendance by 9:48 am. Crews attending the fire reported significant difficulty in initially locating the fire, but a crew entering from the rear of the premises located a significant and fully involved fire in Coo’s kitchen area.

9. At about 10:00 am Mr Hwang tried to return to Coo but was told by police in attendance that he could not return to the restaurant. Mr Hwang told police and fire officers that he may have forgotten to turn the gas off.

10. Firefighters observed the damage to Coo’s kitchen was significant. Of note, one of the fridge/freezers was badly burned compared to the other two fridge/freezers situated either side of it; specifically the compressor part of the fridge was completely destroyed

and the other two appliances were not as badly damaged. The exhaust fans were operative and likely on at the time of the fire. This meant the fire was able to build and grow within the kitchen for a significant period of time without spreading, during which time it is likely a hot layer was able to develop within the kitchen above the range hoods, allowing the room to reach a point where it has flashed over and the fire has travelled quickly outwards in search of oxygen, causing a pressure wave which forced the front door and windows open. Once the fire had this new oxygen supply it was able to build and spread, travelling up through the first floor and into the roof space from where it was able to travel sideways through the brick dividing walls along the timber roof beams.

11. Firefighters had difficulty suppressing the fire, and a lack of modern active and passive fire suppression systems aided the spread of the fire to adjacent tenancies, with multiple tenanted premises in the building being damaged either directly by the fire and/or by the fire suppression methods utilised.

12. The fire was deemed under control by 4:18 pm. Progressively over that day and the next day the public restrictions to the area were lifted and premises released to occupiers.

13. The initial view of ACTF&R was that the cause of the fire was undetermined but not suspicious. After investigation, the area of origin of fire was considered by both AFP Forensics and ACTF&R to be Coo’s kitchen. A specific point of origin could not be determined due to the kitchen having experienced a flashover and extremely high temperatures and fire damage. The Government Gas Inspector excluded the cause of the fire as being a gas leak. The Gas Inspector considered there were two possible scenarios for the cause of the fire, being either the compressor in the bottom of the middle fridge, or the result of a gas appliance being on which in turn ignited the cooking oil inside the wok. Advice from the ACT Planning and Land Agency Electrical Inspector to ACT F&R was that because of the significant damage to all electrical appliances, he could not determine electrical ignition from any appliances in the room of origin, but the extensive damage to the middle fridge indicated that this could not be excluded as the cause of the fire. WorksafeACT concluded that there were no evidence of offences committed under or otherwise breaches of the Work Health and Safety Act 2011.

14. ACTF&R advised me that in the light of Mr Hwang’s evidence that it was their opinion that it was likely that the oil in the wok overheated and ignited causing the fire.

ACTF&R noted that studies show that auto-ignition of corn oil can occur at 234-254 degrees for virgin oils after 25 minutes, and significantly less if the oil has been reused.

ACTF&R stated that although the significant damage to the middle fridge pointed to this also being a possible source of the fire, they considered it more likely that the cooktop (heat source) and the oil (unsupervised flammable substance) was the source of the fire.

15. In the light of section 55 of the Coroners Act 1997, I directed that Mr Hwang be consulted prior to the issue of my interim findings and invited to make submissions. No

submissions were received.

Matters of Public Safety

16. On 21 February 2014 officers from the ACT Environment and Sustainable Development Directorate (as it then was) (“ESDD”) Gas Inspectorate attended the fire site at about 9:50 am. They took steps to ensure that the site was safe by checking the gas supply to Coo and the surrounding shops was turned off and made safe. The officers found that the gas meter for Coo was turned off but not disconnected and capped off so as to be made safe. The risk identified by these officers was that the main control valve to that gas meter, which was accessible to any person entering the site, could have been turned on and charged the system with gas again. The fire damage to the gas infrastructure meant that a gas leak may have been possible, and given the multiple electrical sources of ignition, may have caused another fire. Those officers also found that the gas meters to the surrounding shops were still in the on position (ie. had not been turned off). A firefighter in attendance at that time told the ESDD officers that he called ActewAGL to come out on the day of the fire to disconnect the utilities, and had assumed that the gas had been made safe as part of that process.

The ESDD officers immediately called the gas distribution contractors to come out and disconnect and cap off the gas supply. The disconnection occurred at about 11:30 am that day. While waiting for the gas contractors to arrive, the ESDD officers turned off and danger tagged all the gas meters around Coo to make them safe.

17. This evidence gives rise to two areas of concern as matters of public safety, as follows: a. The content and efficacy of procedures for isolating or otherwise making safe gas supplies after a fire; and b. The content and efficacy of procedures for post-fire release of multi-tenanted buildings or complex tenancies.

18. The Coroners Office has had various exchanges with ActewAGL and ACTF&R to determine what took place on the day of the fire and about various practices and procedures.

Isolating Gas Supplies

19. In the present case there is no dispute that there was a need to isolate gas at the point of entry to the building at least to Coo and surrounding sites while firefighting activity proceeded and while the site remained unsafe. I consider on the facts of this case that the point of entry gas meter should have been disconnected or capped off as soon as practicable.

20. I sought comment and advice from ActewAGL and ACTF&R about their standard fire response processes. Both agencies accept that: a. ActewAGL does not act unilaterally to disconnect utilities.

b. The onus is on emergency services to make a request to that effect to ActewAGL, to which ActewAGL (or its contractors) will respond.

21. I am satisfied that the practice just described is, in principle, an appropriate one.

22. ActewAGL internal processes require that separate requests be placed in respect of each of electricity and gas, and that the requests be directed to different contractors – the electricity distributor and the gas distributor being different entities. It is agreed by ActewAGL and ACTF&R that there was no request made by ACTF&R on the day of the fire to disconnect the gas supply.

23. ACTF&R acknowledges that a critical part of its risk mitigations activities at emergency incidents is ensuring the isolation of utilities, including gas and electricity, to buildings. I have been advised by ACTF&R that it has, since this fire, updated its Operational Guideline that details command and control procedures at incidents. Additionally, ACTF&R Commanders have been specifically briefed on the need to ensure appropriate isolation is undertaken at multi-tenancy buildings.

24. Although the potential risk to public health and safety resulting from the failure to correctly isolate gas to the Sydney Building was significant, no ongoing matter of public safety arises in light of the advice I have received from ACTF&R about the steps already taken in this regard. In the circumstances I make no further recommendation.

Release of Premises

25. Noting that in critical incidents there is often a level of confusion and a number of actors not necessarily working in coordination, a single agency should take responsibility for ensuring site safety (or responsibility for advising that the site is unsafe) before the site is released to the occupant and/or other public access is permitted. In my view, ACTF&R is best placed to fulfil that role. I am advised by ACTF&R that it has documented processes for release of single tenanted premises such as residences. In those cases there is a template “Release of Premises” (ROP) form which is filled out and signed off both by a fire officer and the person to whom the residence is released. There are however obvious difficulties for post-fire release of multi-tenanted buildings or complex tenancies. The standard “Release of Premises” form does not expressly address the issue of utility connection, but ACTF&R advise that hazards such as compromised gas or electrical installations are routinely included on this form to advise occupiers of potential risk. In any case, the standard form was not completed for this fire because the building was not to be released to a single identifiable person.

26. I am advised that ACTF&R recently reviewed its ROP form and “confirmed that the form remains necessary and the current content of the form remains valid because:  It protects the Territory as it reduces the likelihood that ACTF&R would be responsible for something happening after our resources have departed the incident; and  It also serves as a documented warning to the person receiving the form about the known hazards and dangers associated with the premises.”

27. The ACTF&R response to me also included the following:

“The review also specifically addressed the use of the ROP form for multi-tenancy buildings and concluded that it is necessary for the ACTF&R and that the property type and the ownership/tenancy circumstances will determine the most appropriate approach to be taken at the time. This requirement has been communicated to all ACTF&R Commanders, who are now aware of the importance of the ROP form and its use where a multi-tenancy building is involved in a fire.”

28. It is crucial that responders direct their attention to potential risks when premises are released after an incident, and that occupants are fully informed of such risks at the time of release. The ROP form is a useful mechanism for this. In my view it should expressly: a. Include reference to the status of utilities such as electricity and gas etc.; and b. Prompt consideration of identifying the appropriate persons to be made aware of risks in the case of multi-tenancy buildings.

I recommend accordingly.

Fires in Heritage Buildings

29. Information received in the course of this inquiry is to the effect that there are no official requirements for implementation of Fire Management and Suppression Systems in ACT heritage buildings such as the Sydney Building. Rather, the ordinary requirements for fire protection apply equally to heritage buildings, based upon considerations such as the requirements at the time of original construction and any subsequent work.

30. I make no recommendation in this area.

Copies of Findings

31. I direct that a copy of my findings in this matter be sent to the Minister for Emergency Services, the Minister for Planning and the Directors-General of the relevant the ACT Directorates for their information. I acknowledge the cooperation of ACT Fire & Rescue, the ACT Environment, Planning and Sustainable Development Directorate and ActewAGL.

32. I direct also that copies of my findings be made available to any persons whose property was damaged by this fire (or their representatives) upon request made to Court staff.

33. I direct that a copy of my findings will be uploaded in due course to the Court website.

P.J. Morrison Coroner