Response to elevated levels of carbon monoxide in the community


Dr Lester reported to the Board that from 12 February 2014 there was concern, including among the community, about the levels of carbon monoxide being produced by the Hazelwood mine fire.51

On 13 February 2014, the EPA and CFA conducted monitoring of the carbon monoxide levels in the community.52 Dr Paul Torre, Science officer at the EPA, advised that no elevated or significant readings of carbon monoxide were obtained from testing undertaken in the community on 13 or 14 February 2014.53

Mr Costa Katsikis, MFB HazMat technician and Deputy Incident Controller, reported to the Board that on 15 February 2014, CFA HazMat technicians recorded elevated readings of carbon monoxide in the Morwell community. The carbon monoxide spot readings were elevated in and around Morwell, with a peak reading recorded near the Morwell Police Station, which is located south of Commercial Road.54 Mr Katsikis stated that a meeting was promptly held with the Incident Controller, Scientific Advisor, and the Public Information Officer, who agreed that a ‘shelter in place’ warning should be issued to local residents in the affected area. Consequently, a ‘Watch and Act’ alert was issued by the CFA via text message, to a defined group of residents located close to the mine, at approximately 1 pm (see Figure 4.50).55 The message stated: ‘Watch and Act: Morwell residents indoors immediately, close windows/doors/vents. Seek further info via radio.’56

Figure 4.50 Distribution of ‘Watch and Act’ alert on 15 February 201457


Mr Craig Lapsley, Fire Services Commissioner, informed the Board that the national emergency alert telephone alerting system (developed after the 2009 bushfires) was used to send the ‘Watch and Act’ alert to more than 26,000 fixed and mobile telephone subscribers within or passing through the area.58

Dr Lester told the Board that she was advised by the health staff at the Regional Incident Control Centre of the intended notification and sought to provide a risk assessment to the Incident Controller.59 In response to the request for a copy of the risk assessment, the Board was provided with an email to Dr Lester from Mr Julian Meagher, Manager Public Health Emergency Management, Office of the Chief Health Officer. The email was sent at 3 pm on 15 February 2014 (after the ‘Watch and Act’ alert was issued). The email stated that:

…the Incident Controller used a determination that is for exposures in HazMat incidents of 9 ppm however this trigger is an occupational one and is on the basis of 9 ppm over an 8 hr period… the level of 15 ppm is one that a normal person would be exposed to from heavy traffic… or indeed from cooking dinner over a gas stove.60

The Board was not advised whether this information was also provided to the Incident Controller and if so, if the information was provided before or after the ‘Watch and Act’ alert was issued.

The Department of Health was not involved in the decision to issue the ‘Watch and Act’ alert.61 Dr Lester told the Board that she did not agree with the ‘Watch and Act’ alert being distributed and considered that it was unhelpful as it sent a very concerning message to the community that was not necessary.62

Later that afternoon there was an easterly wind change, which dispersed the carbon monoxide.63 The ‘Watch and Act’ alert was downgraded at around 6.45 pm and residents were sent a further text message that stated: ‘Watch and Act – can go outside and open doors and windows.’64

Mr Katsikis informed the Board that the CFA continued to measure increased levels of carbon monoxide on 16 February 2014. The readings on this day averaged 20-30 ppm, with a peak of 60 ppm. Readings were taken inside and outside the Morwell Bowling Club (South) and were consistent for a number of hours.65

The Traralgon Incident Management Team was aware of the consistently high carbon monoxide readings and considered actions ranging from evacuation to a community warning. A ‘shelter in place’ warning was discussed, however the feasibility of this was dismissed as the readings from the Morwell Bowling Club (South), both internally and externally, indicated that sheltering in place would still pose an exposure risk.66

Mr Katsikis stated to the Board that later that day he was informed that an agreement had been reached in relation to community warnings, and that the EPA would provide the information to the Department of Health, which would ultimately decide on the appropriate community warning to be issued.67 Dr Lester told the Board that she was not involved in this decision. However, it was her understanding that the Incident Controller cannot divest himself or herself of responsibility for communications.68

On 16 February 2014, the EPA, with the CFA, continued to monitor the levels of carbon monoxide in the community.69 The readings were reported to the Department of Health via an email from Dr Torre to Ms Vikki Lynch, Advisor, Health Risk Management, Department of Health, at 8.41 pm. The email stated that from 12.30 am to 8.30 am that morning, carbon monoxide levels ranged from 25 ppm to 45 ppm. The email also included a number of short-term readings at various locations in Morwell taken from 1.30 pm to 6.30 pm. These readings ranged from seven ppm to 57 ppm. The Department of Health determined that no further action was required that evening. The email makes reference to a discussion between Ms Lynch and Dr Torre that occurred prior to the email, however no evidence has been provided to the Board about this discussion.70

Dr Lester told the Board that whilst she considered that the readings were high, which was concerning, she understood that the readings were spot readings and so were not sufficiently reliable to inform public health advice or to trigger the Carbon Monoxide Response Protocol (discussed below). She advised the Board that to her knowledge, Ms Lynch did not utilise the data to calculate an indicative average of the readings.71

No warning or advice was issued to the community in relation to high carbon monoxide levels in the southern part of Morwell during the afternoon and evening of 16 February 2014.

Dr Torre told the Board that the levels of carbon monoxide recorded in the community even surprised the EPA:

… those carbon monoxide levels, were very unusual. I’ve never seen carbon monoxide levels at that concentration – not that I’ve seen a lot of coal mine fires, but I was really surprised at the elevated levels. Even when we tried to do a correlation between the particle levels and carbon monoxide, we couldn’t find a pattern. It was really such a different fire. Carbon monoxide levels I’ve never seen before.72


Dr Lester told the Board that given the high levels of carbon monoxide recorded in the community, on 15 February 2014, the Department of Health developed the Carbon Monoxide Response Protocol to provide a decision-making tool to assess the risks to the community of the elevated levels, and if necessary provide advice to the Incident Controller.73 The Department of Health’s Principal Health Risk Advisor and Air Quality Specialist drafted the protocol, together with a number of other relevant medical and environmental health professionals within the Department.74

The National Environment Protection Measure (NEPM) for Ambient Air Quality provides that the ambient air quality standard for carbon monoxide for an average period of eight hours is 9 ppm, and that this should only be exceeded on one occasion per year.75 Dr Lester told the Board that the NEPM was not an appropriate standard for the Carbon Monoxide Response Protocol, as it was intended for longer periods, not for an acute event.76 Dr Lester explained that the ‘Protective Action Decision Guide for Emergency Services during Outdoor Hazardous Atmospheres’, signed off by all relevant Victorian agencies in 2011, was used as the basis for the selection of the thresholds for the Carbon Monoxide Response Protocol.77 The Guide recommended that the Acute Exposure Guideline Levels (AEGL) be used for short-term community exposures to outdoor air chemical concentrations for a range of hazards. The Carbon Monoxide Response Protocol was based on AEGL-2 levels, which relate to the ‘airborne concentration above which it is predicted that the general population, including susceptible individuals, could experience irreversible or other serious long-lasting effects.’ The AEGL-2 aims to maintain a person’s carboxyhaemoglobin level at less than four per cent.78

Figure 4.51 Health protection air levels according to AEGL-2

Health protection air level for
carbon monoxide (ppm)

Average period for monitoring against the
health protection level for carbon monoxide

AEGL–2 levels for carbon monoxide:

  • 420 ppm
  • 150 ppm
  • 83 ppm
  • 33 ppm
  • 27 ppm
Averaged over:

  • 10 minutes
  • 30 minutes
  • 1 hour
  • 4 hours
  • 8 hours

The Department of Health used the levels indicated in Figure 4.51 above to develop the Carbon Monoxide Response Protocol. The Department of Health then applied a further level of conservatism by lowering the standard for one hour exposure from 83 ppm to 70 ppm.79

The Carbon Monoxide Response Protocol provides that if the one hour average value of 70 ppm carbon monoxide is reached, or if the four hour average value of 33 ppm carbon monoxide may be reached, or if the eight hour average of 27 ppm carbon monoxide may be reached, then:

  • The Department of Health and the Regional Deputy Commander must be immediately advised.
  • The Department of Health will convene an internal assessment team to assess the data against the AEGL-2 advice matrix (see Figure 4.52) continuously until the situation is resolved. The advice matrix recommends a number of different messages to the community depending on the level and duration of the exposure to the plume, including recommendations to ‘watch and act’, to ‘shelter in place’ and to relocate.
  • The EPA will verify the results by continuous monitoring of carbon monoxide levels averaged over 15 minute periods and short-term spot monitoring (including deriving the area of smoke plume and wind direction).
  • If the results confirm that the levels exceed the one hour average or meet the four or eight hour average trigger level carbon monoxide reading, the Department of Health will advise the Regional Incident Controller who will convene a Regional Emergency Management Team to deploy specialists to identify the safest area for potential redeployment of the community.80

The Carbon Monoxide Response Protocol also includes an advice matrix to assist the Department of Health’s assessment team when considering advice to the Incident Controller (see Figure 4.52). The carbon monoxide readings in the table below are hourly averages, not spot readings. It is unclear why the Carbon Monoxide Protocol advice matrix includes the level for 83 ppm (consistent with the AEGL-2 guidelines) and not 70 ppm for the one hour exposure standard.

Figure 4.52 Carbon Monoxide Response Protocol advice matrix81

CO Readings Matrix

Predicted Duration of Plume
(How long the plume is in the area)
CO Readings ppm >12 10–12 8–10 6–8 4–5 2–4 1–2 <1
EWSIP – (Emergency Warning Shelter in Place)


  • Shelter in place provides 6 hour protection before the equalisation with the external atmosphere
  • CO based on average reading over a 30 to 60 minute period
  • BoM to provide meteorological forecast of wind speed, direction and duration
  • BoM prediction to inform the estimated time of exposure
EWEVAC – (Emergency Warning Evacuate)
WSIP – (Watch & Act Shelter in Place)
A – (Advice)
All Clear

The Carbon Monoxide Response Protocol applied significantly higher carbon monoxide exposure standards for the community than those applicable under the Health Management and Decontamination Plan, which applied to firefighters (discussed in Chapter 4.4 Firefighter health). The Victorian Government submitted that the justification for this difference was that firefighters are in an environment that is much more exposed to the hazard and that carbon monoxide levels would be expected to dissipate more rapidly in the open air outside the mine.82

Dr Lester informed the Board that from 19 February 2014, when the Department of Health started receiving validated carbon monoxide data from the EPA, the levels of carbon monoxide did not indicate any potential risks to public health.83 There was no evidence before the Board that the Carbon Monoxide Response Protocol was triggered during the Hazelwood mine fire.

Peer reviews of the Carbon Monoxide Response Protocol

The Department of Health and the EPA each had the Carbon Monoxide Response Protocol independently peer reviewed in late February 2014. The peer reviews raised a number of concerns about the Protocol.

The Department of Health had the Carbon Monoxide Response Protocol reviewed by Ms Lyn Denison, Principal Scientist at Toxikos. Ms Denison concluded that the Protocol was appropriate, however commented that if fires lead to prolonged periods (days to weeks) of consistently elevated exposure to carbon monoxide, the AEGLs are not appropriate and the triggers would need to be revised to reflect ‘sub-chronic’ rather than acute exposure.84

The EPA had the Carbon Monoxide Response Protocol reviewed by epidemiologists Dr Fay Johnston, Senior Research Fellow, Environmental Epidemiology, Menzies Research Institute, Tasmania, University of Tasmania, and Professor Ross Anderson, Professor of Epidemiology and Public Health, St Georges University of London and Kings College London. Dr Johnston and Professor Anderson both raised concerns about the use of the AEGL-2 guidelines as the response framework for the Protocol.85 The AEGL-2 guidelines are designed to protect against carbon monoxide exposures of a concentration and duration that would be expected to produce carboxyhaemoglobin of four per cent. Both Dr Johnston and Professor Anderson advised that recent research suggests that there is a wide range of adverse health effects at lower carboxyhaemoglobin concentrations. They suggested that the Protocol should aim to produce carboxyhaemoglobin of less than two per cent, consistent with the World Health Organisation’s guidelines.86

Dr Johnston further suggested that the Carbon Monoxide Response Protocol needs to clarify the use of procedures for the direction ‘shelter in place’ and community relocation in an emergency. Dr Johnston noted that effective protection of the community assumes rapid receipt of information by the community at any given time of the day or night, and pointed out the following issues of concern in relation to this:

  • the extent of community confusion over repeated advice to commence and terminate
    ‘shelter in place’ advice
  • the age of the housing stock in the Morwell area meant that houses may not provide sufficient protection from carbon monoxide in the event that it is recommended that residents ‘shelter in place’
  • the need to clarify the detailed procedures for a possible relocation of the community in
    an emergency
  • a direction to ‘shelter in place’ may not be appropriate for increased levels of carbon monoxide exposure due to the unpredictable timing and duration of the carbon monoxide emissions.87

Professor Campbell advised the Board that he agreed with the reviewers that a lower level of carbon monoxide as the threshold would have been preferable and that as a clinician he would have preferred a margin of protection for vulnerable groups.88

Dr Lester was unable to confirm whether or not the Department of Health was provided with the peer reviews obtained by the EPA.89 However, Mr Merritt told the Board that it was safe to assume that the EPA peer reviews of the Carbon Monoxide Response Protocol were provided to the Department of Health.90 Dr Torre said that he assumed that the peer reviews were passed onto the Department of Health.91 The Victorian Government submitted, after the conclusion of the hearings, that the review by Ms Denison included consideration of the comments from Dr Johnston and Professor Anderson.92 However, there is no evidence before the Board to support this submission.