Coroners Act 1995
Coroners Rules 2006
Rule 11

I, Glenn Hay, Coroner, having investigated the death of

Rodney Alan Williams, Matthew Keith Williams and Terry Charles Bartle

have decided not to hold a public inquest hearing into their deaths because my investigations have sufficiently disclosed the identities, the time, place and causes of death, relevant circumstances concerning how the deaths occurred and the particulars needed to register the deaths under the Births, Deaths and Marriages Registration Act 1999. I do not consider that the holding of a public inquest hearing would elicit any information further to that disclosed by the investigations conducted by me.  Pursuant to section 50 of the Coroners Act 1995 I have investigated the three deaths in the one inquest.

I Find that:

1. (a) Rodney Alan Williams (Mr R. Williams) died on 17 March 2012 at Top Marshes, near Miena in the Central Highlands.

(b) Mr. R. Williams was born in Tasmania on 28 January 1959 and was aged 53 years at the time of his death. Mr R. Williams resided at Invermay with his fiancé and was employed as a Line Supervisor. He had children from a previous relationship.

(c) Mr. R. Williams died due to accidental carbon monoxide toxicity while sleeping in a caravan in which a portable gas-operated refrigerator was in operation.

2. (a) Matthew Keith Williams (Mr M. Williams) died on 17 March 2012 at Top Marshes, near Miena in the Central Highlands.

(b) Mr. M Williams was born in Launceston, Tasmania on 29 March 1985 and was aged 26 years at the time of his death. Mr M.Williams was the son of Mr R. Williams and resided in Moonah with his defacto partner and their 12 month old son.  Mr M. Williams had children from a previous relationship and was employed as a painter.

(c) Mr. M. Williams died due to accidental carbon monoxide toxicity while sleeping in a caravan in which a portable gas-operated refrigerator was in operation.

3.  (a) Terry Charles Bartle (Mr Bartle) died on 17 March 2012 at Top Marshes, near Miena in the Central Highlands.

(b) Mr. Bartle was born in Launceston, Tasmania on 6 October 1972 and was aged 39 years at the time of his death. years.  He resided in Mowbray with his wife and four year old son and was a ‘stay-at-home’ dad by occupation.

(c) I find that Mr. Bartle died due to accidental carbon monoxide toxicity while sleeping in a caravan with a portable gas-operated refrigerator was in operation.

Findings relating to the circumstances surrounding the deaths:

Mr R Williams was the father of Mr M Williams while Mr Bartle and Mr R. Williams had known each other for many years, with Mr Bartle treating Mr R Williams as an older brother. They were all keen shooters of deer and they hunted within a small group of similar-minded friends which had, over the previous two years, included Michael Jon Wilson (Mr Wilson). Mr Wilson joined the group when invited and after satisfying himself that the others were responsible gun owners and users.  Other witnesses gave evidence of the competence and the concerns for safety displayed by all members of the group.

As a result of a successful ballot the group was able to shoot in a conservation area at Top Marshes, near Miena in the Central Highlands on the weekend of 16, 17 and 18 March 2012.  All four persons travelled to the area and met approximately two to three kilometres from the area known as Top Marsh Reserve late in the afternoon of Friday 16 March. They set up camp there. Mr Wilson was accompanied by his nine year old son, having previously satisfied himself that it was a safe environment for him.

Mr Bartle, Mr M. Williams and Mr R. Williams agreed to sleep in a caravan provided by Mr Bartle while Mr Wilson chose to sleep with his son in a nearby tent that he had brought with him.

The caravan had been purchased by Mr Bartle two years previously and he spent some time in refurbishing it.  He had not refurbished a caravan previously but is said to have been handy at fixing things generally.  He stripped it to the shell of the caravan frame as the roof had been leaking.  He relined it fully and built all of its furniture and fixtures.  He placed new seals around all doors and windows and other outlets to the outside. 

During the evening of Thursday 15 March Mr Bartle sent a text message to Mr Wilson “Hey buddy could you bring a tarp up with you the caravan roof is leaking and need to cover it. “

The caravan had an inbuilt refrigerator which ran from the external main gas cylinder.  There was also a portable Jackeroo refrigerator which could run from both gas and electricity.  It was usual to have the portable fridge outside the caravan hooked up to the main gas cylinder.  There is evidence from Mr Bartle’s wife that the portable fridge was never used inside the caravan but was packed between the bunk bed and the kitchen bench during transport.

At about 7.00pm on 16 March Mr Bartle telephoned his wife, complaining that things had gone wrong at the camp site, in that they had not yet seen any deer and the pilot light on the refrigerator was playing up.  It is not known which fridge he was referring to.  He also said that the generator was not working as the spark plug could not be changed.  Further, he complained that the portable gas hot water heater in the rear of the utility had caught fire.

Around 10.00pm on 16 March there was a heated verbal exchange between Mr Bartle and Mr R Williams over the failure of some of the equipment that Mr Bartle had provided for the weekend. When the pair continued to swear in the company of Mr Wilson’s son, he and his son retired to their tent for the night.  Mr Wilson and his son could hear the other three continue to talk around the campfire, until they both fell asleep.  It is clear the three men continued to consume some alcohol around the campfire.

At about 12.15am on Saturday 17 March, Mr Bartle sent a text message to his wife saying that he had fallen asleep by the camp fire.

Shortly after 5.00am on Saturday 17 March Mr Wilson and his son were woken by noises of someone coming out of the caravan for a short time before re-entering and closing the door behind them.  They could hear noises including yelling and banging coming from inside the caravan but what was being said was not clear to them. Mr Wilson identified the voice of Mr Bartle say “Rocket get off me I can’t breathe.”   Mr R. Williams was otherwise known as “Rocket.”  When Mr Wilson’s son became distressed at the shouting and noises he decided that they should leave the area as he was not sure what was occurring inside the caravan believing Mr Bartle may be having some form of a breakdown and was further concerned as he was aware there were firearms in the caravan.  He did not go to investigate the noises as he was scared and concerned for the safety of his son and himself.

At about 5.30am Mr Wilson and his son drove from the area, to Little Pine Lagoon where there was mobile telephone coverage.  At 6.35am Mr Wilson telephoned Mr M. Williams but could get no connection.  At 7.20am Mr Wilson telephoned Mr Bartle’s number; the telephone rang but there was no answer.  At 7.21am Mr Wilson telephoned Mr R Williams but there was no connection.

As it was daylight, Mr Wilson and his son returned to the campsite to check on the others.  Leaving his son in the vehicle Mr Wilson approached the caravan, calling out to the occupants.  When he received no response he cautiously opened the door and saw all three persons laying on the floor and appearing to be asleep. He was concerned as the interior of the caravan was in disarray and so decided to return home. 

After gathering their belongings from the tent Mr Wilson returned to check once again on those in the caravan.  He could not see any indication of any of the three breathing and upon checking was unable to detect a pulse in any of them.  On this occasion he smelt gas within the caravan.  He believed that all three were deceased and immediately packed all his further belongings and drove with his son, both in a distraught state, to the area of Little Pine Lagoon where he rang ‘000’ He advised emergency services that he believed all three persons were deceased and then drove to the Great Lake Hotel at Miena to meet the responding emergency services. Upon their arrival Mr Wilson guided the volunteer ambulance officers to the caravan.  They opened the door to the caravan, immediately smelled gas and observed three bodies lying on the floor.  Subsequent investigations found that Mr M. Williams and Mr Bartle were lying with their heads towards the front and near the door of the caravan and they were partly in their sleeping bags.  Mr R. Williams was lying partly on top of the other two, near the door to the caravan and dressed in a t-shirt and underwear, but not in a sleeping bag.  Just inside the caravan door was a 9kg gas tank which was turned off.  On the far side of the caravan was a portable gas refrigerator on the floor between the kitchen bench and bunks.  This portable refrigerator and an adjacent gas light were connected to a gas bottle.

Ambulance officers partly vented the caravan and confirmed all three persons were deceased.

Police attended the camp site and commenced an investigation on my behalf.   Tas Fire Service also took gas readings inside the caravan and ascertained there were dangerous levels of carbon monoxide present inside the caravan which was further ventilated.

The investigations established there was no indication of violence or the presence of any unknown person.  The caravan keys were insitu in its external door lock and keys to a nearby vehicle were located in its ignition. Firearms lay undisturbed across both the front and rear vehicle seats of that vehicle, where they had been previously left.

A generator was found in the rear of a nearby vehicle and an electric cord plugged into an outlet on the exterior wall of the caravan, but not connected to the generator.  Condensation was noted on the interior of the caravan windows.  There were several gas-operated caravan/camping appliances found inside the caravan.

The examination of a 9kg portable L.P. gas bottle located fixed to the caravan external chassis indicated that this had not been in operation at any relevant time. 

When it was determined safe to enter the caravan a detailed forensic examination was conducted of it -

• Examination of a 9kg portable L.P. gas bottle located directly inside the caravan to the left of the door and connected to a lantern fitting indicated that it was not in operation.
• Examination of the fitted L.P. gas refrigerator beneath the bench in the kitchen area indicated that it was not in operation.
• Examination of a 4.5kg portable L.P. gas bottle located to the right of the doorway and adjacent to and connected to a bench top gas cooker found it was in operation. It was also connected to a ‘Jackeroo’ brand portable refrigerator which was also in operation.  Gas could be smelt in this area of the caravan and it was determined that there appeared to be a leak in the line leading from the bottle to the cooker.  A further examination of the cooker indicated that the burner jets were in the ‘off’ position, however when operated, the gas could be heard to freely flow as intended.
• Preliminary examination of the portable refrigerator indicated that it was in operation and cooling foodstuffs located inside it.  It was clearly labelled “WARNING – WHEN OPERATING ON GAS USE OUTDOORS ONLY.  A dangerous atmosphere may result if gas is used in any enclosed environment including vehicles, caravans, boats, tents or unventilated rooms.”
• All windows, the roof vent and door were completely closed and well-sealed.   Any other openings to the outside were also sealed to the extent that there was no ventilation to the caravan.

Subsequent Investigations relevant to findings:

The caravan and its contents was transported to Hobart and impounded by police investigators for further examination. 

On the morning of 19 March 2012 post-mortem examinations of all three persons were conducted in Hobart by Forensic Pathologist, Dr Donald Ritchey.  As a result of his examination Dr Richey determined:-

• in the case of Mr R. Williams that the cause of death was carbon monoxide toxicity.  Toxicology testing revealing a fatal saturation level of carboxyhaemoglobin of 59%.  There was no indication of any injury to or violence suffered by Mr R. Williams.
• in the case of Mr M. Williams that the cause of death was carbon monoxide toxicity.  Toxicology testing revealing a fatal saturation level of carboxyhaemoglobin of 63%.  There was no indication of any injury to or violence suffered by Mr M. Williams.
• in the case of Mr Bartle that the cause of death was carbon monoxide toxicity.  Toxicology testing revealing a fatal saturation level of carboxyhaemoglobin of 60%.  There was no indication of any injury to or violence suffered by Mr Bartle.

Levels of carboxyhaemoglobin are normally less than 10%.  Levels that exceed 50% saturation are considered life threatening.

I accept and adopt the opinions of Dr Ritchey and also accept the toxicology reports.

While there was evidence of premixed alcoholic drinks at the camp site and that each of the deceased had been drinking earlier in the evening, toxicology testing of blood and urine taken from each of the deceased later on the 17 March did not detect any alcohol present in their bodies.  I can make no finding as to the level of intoxication of each of the deceased at any time prior to their death.

The caravan and fittings were extensively and exhaustively examined by appropriately qualified personnel from Workplace Standards Tasmania.

As a result of this examination it was determined that the caravan had undergone recent extensive renovations with most internal fittings removed, replaced or relocated. This included:-

  • the replacement of the interior lining of the caravan, however this did not include the placement of interior venting to align with the manufacturers pressed-metal vents on the exterior surfaces;
  • the installation of a double thickness foam rubber seal around the door frame;
  • the installation of a new bunk beds, a double-size bed, a kitchen bench, shelving and sink which drained through the floor; and
  • the relocation of a built-in gas refrigerator. This had hard plumbed fittings connected to a 9kg portable L.P. gas bottle externally mounted on the caravan draw bar frame and was vented through the floor however did not have a 'gas fitting compliance tag' fitted.

At the time of the incident there was a portable 9kg L.P. gas bottle with a lantern fitting attached located immediately inside and to the left of the caravan door.  Testing of this appliance revealed no issue with it or that it contributed to this incident.

On the far side of the caravan interior, opposite the door and on the floor between the kitchen bench and the bunk beds was a 4.5kg portable L. P. gas bottle. It was located adjacent to a bench mounted two-burner gas cook top and a portable ‘Jackeroo’ brand refrigerator.  The gas bottle was connected to the refrigerator through flexible gas tubing and a regulator attached to the bottle.  Inspection of the refrigerator indicated that at the time of this incident it was operating by means of the gas supply.

Also attached to this gas bottle was a vertically placed chrome tube. This was connected to flexible gas tubing which, in turn, was fitted to the bench-mounted two-burner gas cooker.  Inspection of the cooker indicated that the burner cut-off valves were in good working order as the gas flow ceased completely when the valves were operated. The inspection did not identify any issue with this appliance or that contributed to this incident.

Inspection of the connecting chrome tubing and flexible hose that provided the gas supply from the bottle to the cook-top revealed a leak where the tubes were joined.  While the connecting fitting had an isolating valve fitted and was operational to prevent the flow of gas to the cook top, the leak was prior to the isolating valve and so the release of gas was constant while the bottle-mounted valve was open.  The bottle-mounted valve remained open at all times as there was a need to have the portable refrigerator in use as it shared the gas supply with the cook-top.

The investigation estimated that the occupants of the caravan may have been inside, without adequate ventilation and with gas leaking from the connector fitting for a minimum period of 5.5 hours.  Testing indicated that 22.8 litres of L.P. gas leaked over this period and, being heavier than air, would cover the floor of the caravan to a depth of 3mm. 

Tests were conducted to monitor the carbon monoxide (C0) and oxygen (02) levels within the sealed caravan with only the portable ‘Jackeroo’ refrigerator in operation.  Results establish: -

• The CO level after 75 minutes was 353 ppm (parts per million).  Interpolated results suggest that the CO level of 1200 ppm would have been reached after 4.0 hours. 
• The O2 levels dropped from 20.7% to 19.8% after 4.5 hours.

The Commonwealth Department of Sustainability, Environment, Water, Population and Communities and the National Occupational Health and Safety Commission has published the following information;-

Carbon monoxide quickly enters the blood when inhaled into the lungs. Levels normally present in the atmosphere are unlikely to cause ill effects. Carbon monoxide concentration may reach harmful levels in poorly ventilated rooms during operation of unflued gas heaters or defective non-electric heating appliances, or in the passenger compartment of vehicles with defective exhaust systems. At low levels it may cause poor concentration, memory and vision problems and loss of muscle coordination. At higher levels (200 ppm for 2 to 3 hours), it may cause headaches, fatigue and nausea. At very high levels (400 ppm) the symptoms intensify and will be life threatening after three hours. Exposure to levels of 1200 ppm or greater are immediately dangerous to life.

Safe Work Australia has published information that safe/acceptable oxygen levels in the air as being between 19.5% and 23.5%.

Testing was also conducted on the ‘Jackeroo’ refrigerator and confirmed that the appliance was in good working order.  The exterior of the refrigerator was clearly labelled with warnings it should only be used outdoors, indicating that use in enclosed environments may create a dangerous atmosphere.  Documentation relevant to this appliance confirms this warning and gives clear instructions on the correct and safe use of it.

It was noted from scene photographs that a pillow was located on top of and adjacent to ventilation grates on this refrigerator.  If this was the case at the time of the incident and whilst it was in operation, the position of the pillow most probably would have:-

• Become damp with water vapour and prevented/restricted the release of combustion products;
• Impeded the supply of fresh air for combustion, and/or
• Caused sooting of the gas burner.

Either individually or collectively these issues would have caused elevated levels of CO.

Other Findings and Comments:

Following a detailed investigation I find that there were no suspicious circumstances surrounding the deaths of any of the deceased.

I find that each of the three men died due to accidental carbon monoxide toxicity while sleeping in the caravan while a portable gas-operated refrigerator was in operation and a gas hose fitting was also leaking gas into that sealed caravan.

It is not possible to give final and unequivocal answers to many questions arising from the deaths of these three men.  However, based on all of the available evidence it is plausible that Mr R Williams left the caravan at around 5.00am, possibly to relieve himself, when he was in a carbon monoxide intoxicated state.  That when he re-entered the caravan closing the door behind him he realised something was wrong and attempted to drag his son and Mr Bartle from the bunks in which they were sleeping in an attempt to wake them or drag them from their sleeping bags or from the caravan, but he in turn was completely overcome by the carbon monoxide before he could complete this task.    The positioning of each of the deceased upon finding suggests this was the case. There is evidence from others that the three men had usual sleeping places on bunks in the caravan.

Medical evidence suggests that any of the men, if not unconscious, would have been in confused states due to carbon monoxide toxicity and lack of oxygen.  Muscle co-ordination would have been impaired and that may account for the noises heard by Mr Wilson at around 5.00am.  The symptoms of high level carbon monoxide toxicity prior to death range from impaired mental state, vertigo, ataxia (loss of physical co-ordination due to underlying damage to the nervous system and brain), breathlessness, seizures and then loss of consciousness.

Further, while there is evidence from Mrs Bartle that the portable refrigerator had never been used in the caravan before and Mr Wilson also recalled that on a previous hunting trip the annex of the caravan had been used with all the cooking gear set up there, it is not possible to speculate why on this occasion the portable refrigerator and other equipment was being used inside the sealed caravan.

In my view Mr Wilson took all appropriate steps at all relevant times and in all the circumstances of the tragedy.

For any gas appliance to operate efficiently, the space in which the appliances installed requires adequate ventilation for the ingress of fresh air to the appliance for combustion and ventilation for egress of combustion products. Inadequate ventilation will starve the appliance of oxygen required for stoichiometric (complete) combustion. Incomplete combustion produces carbon monoxide.

The Gas Act 2000 and the Gas (Safety) Regulations 2002 mandate compliance with Australian Standard AS 5601 relating to gas installations in caravans.

AS/NZS 5601.2 is prescriptive about the size, location and construction of ventilation openings in caravans to provide minimum free air total permanent ventilation.

The caravan in this case  had permanent vents pressed into the external aluminium cladding but unfortunately these vents did not align with any ventilation openings in the lining on the inside of the renovated caravan, preventing any natural ventilation. All windows and the door appeared to have new self -adhesive foam sealing strips fitted, most likely during the caravan renovation by Mr Bartle. Further, there were two penetrations from the inside to the outside of the caravan, one for the exit it of the flexible flue on the inbuilt refrigerator, this was a tight fit with very little air gap between the flue and the penetration. The second penetration was for the entry of the gas supply for the inbuilt refrigerator and other services including sink waste-out. However this penetration had been sealed around the pipes with what appeared to be wetted toilet tissue. The roof vent was found to be in the closed position and had the appearance of having not been opened for some time.

Based on this available evidence of the caravan with window, door and a roof vent closed it is clear it was extremely airtight. It is clear that by the three men occupying the sealed caravan for a period of approximately 5.5 hours, consuming oxygen in their sleep and with the portable refrigerator in the on position also consuming oxygen and producing carbon monoxide and the gas bottle and fitting, leaking that the men died from carbon monoxide toxicity.

The inbuilt refrigerator appears to have been reasonably well installed but there is no evidence of a gas fitting compliance tag, suggestive of an owner installation not certified by an appropriate gas fitter. The camp stove and Jackeroo camp refrigerator and the gas lantern were all outdoor portable gas appliances and not suitable for use inside a caravan, especially non-ventilated.

All of those factors have unfortunately contributed to the cause of these unfortunate deaths. 


During the course of my investigations it became clear that incidents involving significant carbon monoxide poisoning or subsequent death arising therefrom are not uncommon in Australia and elsewhere in the world.  One of my duties is to make recommendations with respect to ways of preventing further deaths or on other appropriate matters.

1. This tragedy gives rise to a timely reminder to all do-it-yourself builders or renovators (not limited to caravans) to properly follow all legislative requirements including necessary safety standards and guidelines.  To do so will limit as much as possible dangers to those persons or others who may come into contact with the items being built or renovated.

2. That all persons use appropriately qualified gas fitters and these gas fitters comply with all relevant legislation and guidelines to prevent such incidents occurring in the future.

3. I adopt the words of a Victorian Coroner in 2000 in handing down findings in a not dissimilar case;-

"Carbon monoxide is produced from the incomplete combustion of burning products and has been aptly described as "the silent killer,” since it is a colourless, odourless, tasteless and a non-irritating gas. It is responsible for a high proportion of poisoning fatalities, many of which could be prevented if proper education and precautionary measures were undertaken."

On 3 March 2011, the Federal Parliament passed a motion that requested the Standing Council on Energy and Resources (SCER) to explore options to mitigate the risk of carbon monoxide (CO) poisoning in residences. A caravan suitable for accommodation is clearly a residence.

In response to this request, SCER developed the Gas Appliance (Carbon Monoxide) Safety Strategy. Expert advice was obtained from the Gas Technical Regulators Committee and the Allen Consulting Group. SCER found that certain gas appliances pose more risks than others and a number of feasible options were canvassed including: a public awareness campaign; mandatory installation of CO alarms; increased ventilation for gas appliances; having motorised ventilation attached to exhaust fans; and improved engineering to natural draught appliances.

SCER found the best option was to increase public awareness of the situations in which those riskier appliances may cause CO poisoning. It decided that State and Territory governments would examine their existing public awareness initiatives for carbon monoxide poisoning and identify opportunities for improvement. The Director of Gas Safety, Tasmania has endorsed the recommendation.

A further main focus for the strategy to this point has been the issue of air tightness of buildings and the adverse effect on natural draught conventionally flued gas appliances.

In light of this triple fatality I adopt and endorse the recommendation of the Director of Gas Safety, Tasmania to expand the scope of any “public awareness” campaign to include portable gas appliances and warnings to users of gas in confined spaces to ensure that gas appliances are safe, properly installed and maintained. I go further and recommend this be done urgently together with urgent consideration in Tasmania of the mandatory installation of CO detection alarms in any residences, boats or caravans where gas appliances are either permanently fixed or where portable gas appliances might likely be used.

4. That consideration is given to empowering regular inspection of caravans or boat gas installations to ensure safety of occupants and to monitor appropriate installation and maintenance of gas appliances and the appropriate ventilation thereof.

5. The Director of Gas Safety, Tasmania alert insurance companies involved in providing protection for licensed gas fitters and the owners of caravans or boats about the risks associated with non-compliance with relevant regulations and Australian Standards surrounding the installation or use of gas appliances in those environments.

6. The Director of Gas Safety, Tasmania alert any plumbing apprentice courses and TAFE and Polytechnic courses required for the registration of gas fitters, to include reference to and testing about the requirement to check for flue operation and appropriate ventilation operation when performing routine maintenance on gas appliances in caravans or boats; to give warnings to owners and to notify the Director of Gas Safety and the Registrar of Motor Vehicles when the unit does not comply with current regulations and standards.

The sad consequence of all of these unfortunate circumstances is that three families and our community have tragically been robbed of loving and vibrant men, so greatly missed by their families and friends.  I wish to express my sympathy and sadness to those families and friends.

DATED: 14 January 2013 at Hobart in Tasmania.


Glenn Hay