Department of Labour v Pike River Coal Limited DC Greymouth CRN 11018500202 [2013] NZDC 807 (3 May 2013)

The prosecution proceeded by way of formal proof. Although both witnesses were present and gave oral evidence in support of their sworn affidavits, there was no cross examination. I did ask questions by way of clarification.

The findings I have made are based solely on the evidence presented before me. I have not read or had referred to me the findings of the Royal Commission on The Pike River Coal Mine Tragedy, or any evidence that was taken before it.

This decision reflects a finding against the Company. It does not assess the culpability of any other person or entity. The decision records an accumulation of errors and omissions which transpired over a number of years. I found a systemic failure of the company to implement and audit its own (inadequate) safety plans and procedures.

The nature of the evidence is technical and has its own vocabulary. To understand the judgement it is necessary to refer to the mine plan Appendix 1 and the Glossary Appendix 2. I have refrained from trying to explain the terms in the judgement as this would only add unhelpfully to its length.

PREFACE ................................................................................................................ 2

Contents.................................................................................................................... 3

Background .............................................................................................................. 7

THE MINE ............................................................................................................... 8

THE CHARGES .....................................................................................................11

CRN 202 and 205 – Methane explosion management ........................................... 12

	Home \| Databases \| WorldLII \| Search \| Feedback District Court of New Zealand

Practicable step 1................................................................................................... 12

Practicable step 2 (pre-drainage of coal seam) ..................................................... 14

Practicable step 3 – Methane drainage range ....................................................... 15

Practicable step 4 – Adequacy of gas sensors ....................................................... 16

Practicable step 5 – Inadequacies in relation to identifying reporting and investigating and resolving causes of gas plugs .................................................... 20

Practicable step 6 – Tube bundling........................................................................ 22

Practicable step 7 – Over pressure detection system ............................................. 23

Practicable step 8 – Outburst potential in the seam .............................................. 23

sensors .................................................................................................................... 24

Practicable step 10 – Identification and investigation methane drainage inflow

and output rates...................................................................................................... 25

Practicable step 11 – Risk assessment for the secondary extraction panel ........... 26

Practicable step 12 – Boart long year LMC75 directional rig .............................. 29

Practicable step 13 – Daily maintenance inspections of the fans.......................... 30

Panel geology, information’s 206 and 209............................................................. 31

maintenance of a ventilation control device (VCD) .............................................. 33

2; PRCL should have ceased mining to allow for the risk of unpredictable but foreseeable events .................................................................................................. 37

3 – Ventilation changes prior to panel move .......................................................... 38

4 – Adequacy of oversight of underground ventilation arrangements ................... 40

Information’s 203 and 207 - Mitigating the risk and impact of explosion............. 43

Practicable step 1- Adequacy and function of explosion protection for the surface infrastructure.......................................................................................................... 43

Practicable step 2 – Placement of the main fan..................................................... 45

Practicable step 3 - Re-evaluation of risk after fan implementation ..................... 46

Practicable step 4 – Minimisation of the risk and damage to the underground fan48

Practicable step 5 – Water or stone dust barriers.................................................. 48

Practicable step 6 – Smoke lines............................................................................ 49

Practicable step 7 – FAB ........................................................................................ 50

Practicable step 8 – PRCL should have implemented an effective system to know the whereabouts of all persons ............................................................................... 51

CRN 211 – Ensuring the safety of Joseph Dunbar................................................. 53

APPENDIX 1 ......................................................................................................... 55

APPENDIX 2 - Glossary ....................................................................................... 57

[1] On 19 November 2010 at 15.45 hours there were the first in four explosions at Pike River Coal Mine. Of the 31 men underground only two survived; 29 men lost their lives as a result of the first explosion.

[2] To date efforts to recover their remains have been thwarted by the hazardous conditions within the mine.

[3] As a result of the explosions, a large investigation was conducted by the Department of Labour, now the Ministry of Business Innovation and Employment (the “Department”).

[4] As a result of the investigation by the Department, the defendant company faces nine charges. The nine charges cover breaches under s 50 and s 18 of the Health and Safety in Employment Act 1992. They allege that the defendant company failed to take practicable steps in three main areas; methane explosion management, minimising the risk of explosion, ventilation management, and lack of investigation into panel 1, 1W1R geology.

[5] The charges fall into three distinct groups and cover the three types of persons who were underground at the time of the explosion, employees, contractors and Mr Dunbar, a new employee of one of the contractors. The employees were:

There was also a young man who was due to start as a contractor for VLI Drilling

[7] On 31 July 2012, the defendant company indicated that it intended to take no part in the proof process; however, it reserved its position with regard to any sentencing which might follow. Accordingly, on 14 and 15 of March 2013, the matter has proceeded by way of formal proof.

[8] The informants presented two affidavits. One from Jane Isabella Birdsall, who also gave oral evidence. The other affidavit was from David Harold Reece, who also gave oral evidence. Mr Reece was the convener of an expert panel that was commissioned by the informant at the start of its investigation. The expert panel consisted of:

[9] This expert panel had access to documents provided by the New Zealand Police, which included reports from the mine, inspection reports, logs, technical reports, feasibility studies, incident reports and later statements from interviews. The expert report was completed in October 2011 and the panel convener, Mr Reece, gave evidence before the Royal Commission into the Pike River Coal Mine tragedy. Mr Reece’s areas of expertise are in underground coal mining operations, specifically bored and flow development in longwall extraction, and gassy and spontaneous combustion-prone environments; emergency response; risk management and safety and health management systems.

[10] I was greatly assisted by counsel for the informant, and also the Amicus Mr Greig. The evidence presented through the two witnesses was comprehensive and extensive. I compliment both of the witnesses for putting what could be described as very technical and difficult information into a format that was readily understandable and digestible.

[11] On 19 November 2010, Pike River Coal Limited (“PRCL”) was a duly incorporated company carrying out the business of underground coal mining. It owned and operated the Pike River mine at Atarau, which is situated 50 kilometres north-east of Greymouth on the New Zealand West Coast. The mine itself is situated in heavily forested and deeply incised Paparoa Ranges under public land that is administered by the Department of Conservation. The nature of the geography has historically been a barrier to extracting the coal. Assessing the seam and a suitable method of extraction were the two principle problems.

[12] There were many feasibility studies undertaken and a decision was finally made by PRCL to mine the seam using a combination of continuous miners, road headers and hydro-mining.

[13] Hydro-mining is the process of excavating stone or coal with the use of high pressure water and specialised equipment. This was the planned extraction process at the mine. Hydro-mining is practiced in only a small number of countries around the world, although it has been used in other West Coast mines in New Zealand. Hydro- mining is an unusual method of mining outside of New Zealand, and there are few experienced miners and managers available in New Zealand and elsewhere who are familiar with this method.

[14] In the mine, access to the seam was by a 2.3 kilometre tunnel from a surface portal. The tunnel rose up on a 1:11 gradient. The tunnel was began in September

2006 and completed in October 2008. The $20 million coal processing plant was completed in January 2009, with a slurry pipeline from the portal to the plant completed the following month. The first extraction of coal, using hydro-mining, began on 22 September 2010. The main extraction fan for the mine was installed and commissioned on 27 October 2010. By the time of the explosion on 19 November, Pike River Coal was behind in its projected schedule and production tonnage.

[15] On 19 November 2010, only one panel in the mine was technically being extracted, panel 1, 1 west 1 right (panel 1, 1W1R). This panel was worked by four

crews working two 12 hour shifts each day, seven days a week. The mine had a designated area known as the restricted zone where only flameproof or intrinsically safe equipment was permitted to be used due to the potential for flammable gas to be present. The criteria for defining this area are found in the Health and Safety in Employment Underground Regulations 1999.

[16] At the time of the explosion operating in the restricted zone were a number of vehicles. There was a specialised mining vehicle, juggernaut load haul dumper, Anderson coal tram, SMV brumby loader, tug with hydraulic digger attachment and ram car. Methane sensors had been fitted to the coal tram and juggernaut and set to alarm when the methane reached 1% of the terminal body of air. The SMV vehicles did not have methane sensors fitted to them. There were also in the restricted zone, a number of pieces of machinery which ran of electricity. An ABM continuous miner, Warratah continuous miner, Warratah road header, hydro-monitor and drill rig. Other potential electrical ignition sources located in the restricted zone included auxiliary fans (four in operation), distribution boxes (three in operation), two restricted zone substation transformers and conductors, power cables and trialling cables.

[17] At the time of the explosion, 31 men were underground and working within the mine. The exact location of the men is unknown due to deficiencies in the tracking of, not only employees, but also of contractors within the mine.

[18] The mine was known as being a ‘gassy mine’. Methane is released from coal and both methane and carbon monoxide were known and well recognised hazards within the mine. Therefore, the ventilation of the mine was of crucial importance in terms of ensuring the safety of those men working underground. Aligned with this was ensuring that ignition sources underground were mitigated by either the presence of sensors which would shut of electricity and/or diesel to the various parts of the machinery and the constant monitoring of the gas levels in the various headings and workings of the mine.

[19] Methane is a naturally recurring gas that is released from coal during the mining process. It is odourless, tasteless and invisible. It is generally detected by either hand-held or fixed electrical gas sensors.

[20] Methane has a specific gravity of just over half of that of air. As a result it may lie in the roof of headings or collect in cavities unless there is sufficient ventilation quantity and velocity to dilute and remove it. Methane is explosive when it makes up between 5 % and 15 % of the volume of air. Above or below these percentages (or concentrations), methane is not explosive.

[21] As noted above, at the time of the explosions, 16 men were employed by PRCL, 12 of the victims were either employed as contractors or self-employed contractors and one victim, Mr Dunbar, was due to start work at the mine as a contractor to VLI Drilling Pty Ltd.

[22] Although no access has been able to be gained to the mine post the explosions, the experts advising the Department have determined that it is most likely that the explosion involved the ignition of a large volume of methane. The most likely source of this methane is from an area of panel 1, 1W1R. At the time of the explosion, the panel had a void volume of approximately 6000 m3, which

reached almost 100 % concentration away from the ventilated fringe. In addition to the methane already in the void, further methane could have been released during the roof collapse from the freshly exposed coal in a rider seam that was located in the roof strata. The potential reservoir of gas from this sort of release could be very quick and as large as 20,000m3.

[23] The most likely way for this methane to have been displaced was for there to have been a roof fall. The methane that was released then needed to be ignited. The most likely scenario for the explosion is that the methane of high concentration was pushed down the panel road due to roof collapse/fall. Concentrated methane has then mixed with the main intake air after knocking over the stopping (VCD) in cross cut 3

1 west and with the main return air to form an explosive mixture. There were a number of potential ignition sources underground on 19 November inbye the panel,

including electrical arcing from EMC currents generated by the variable speed drives, overheating or malfunction of diesel machinery and the failure of protection on electrically-powered plant.

[24] Post the explosion and consistent with the accounts of the survivors, the modelling that has been done suggests that the explosion occurred well into the mine and inbye the location of the main fan.

[25] Explosion is a well-known hazard in underground coal mining and such tragedies regularly take the lives of workers. The key to controlling these hazards is to always keep in mind that with the right confluence of factors, an explosion can occur. Therefore it is essential to maintain regular risk assessments and to take appropriate steps to lessen those risks.

[26] The informant’s case against PRCL is that it failed to implement Health and Safety measures. The informant also claims these were at times inconsistent and inadequate in relation to the risks presented. This led to a failure to introduce overarching and specific risk controls. I have found that there was almost a complete absence of risk assessment undertaken at crucial times in relation to the development of the mine and in particular, the week leading up to the disaster on 19 November

[27] I have also found that although there were various plans for example ventilation management plan, some of this had not been implemented or updated; some of it had been started but was incomplete. There were also a number of safety management plans which were still in draft form on 19 November 2010. Some of these plans referred to particular positions of control, however, as of

19 November, no persons had been employed or allocated these respective roles. There was no effective auditing process in place. There was no site-wide auditing of the crucial safety aspects in relation to the mine, in particular, the gas sensors and the ventilation control devices.

[28] A total of nine charges have been laid against PRCL. Four charges allege that in respect of four mining areas: methane explosion management, ventilation management, panel geology and mitigating the risk of impact of explosion that PRCL, as an employer, failed to take all practicable steps to ensure the safety of its employees while at work.

[29] A further four charges laid under s 18 Health and Safety in Employment Act allege that in respect of the same four mining areas listed above, PRCL as a principal failed to take all practicable steps to ensure that no employees or contactors or sub- contractors or individual contractors or sub-contractors were harmed while doing work at Pike River Mine. A ninth charge laid under s 15 Health and Safety in Employment Act 1992 alleges that as an employer, PCRL failed to take all practicable steps to ensure that no action or inaction on the part of its employee while at work harmed Joseph Ray Dunbar, in particular with respect to ventilation control devices (VCDs).

[30] The informant accepts that at all times it has the onus of proving all the elements of the event, including the failures to take all practicable steps. Once that onus has been discharged, there is no scope for a defendant to argue that there is any defence of due diligence or total absence of fault. See the decision of Judge Morris in the Department of Labour v Gibson O’Connor Ltd, DC Auckland CRN

[31] The Court can also, if it is satisfied on the evidence before it that additional alternative practicable steps were available but not taken, find a failure on the basis of these additional steps subject to the requirement that the defendant is not being prejudice in its defence.

[32] This last matter is particularly relevant in relation to information ending 211 which relates to Mr Dunbar. Although the practicable steps relied on by the informant in relation to that information relates to ventilation control devices, in my

view, having heard all of the evidence, practicable steps which are relevant in relation to CRN 202 and 205 are also relevant in with respect to Mr Dunbar.

[33] I intend in this judgment to deal with the CRNs grouped together covering the four specific areas which the informant alleges significant deficits in relation to the defendant company.

[34] With respect to information’s ending 202, 205, 204, 208 and 2011, I find there is a direct causal link between the breaches on behalf of the company and the explosions that occurred on or about 19 November 2010. I make this finding to clarify that the deceased men and their immediate family members are victims under s 4 of the Victims Rights Act (2002).

[35] Having heard the evidence, I see these breaches of the Act as being significant and fundamental in relation to the Health and Safety of anyone entering the mine.

[36] The informant charges that PRCL failed to take 12 practicable steps with respect methane explosion management. I find both information’s proved and find that the company failed to protect its employees and contractors from the well recognized and understood risk of methane gas explosion. My reasons follow.

[37] The company failed to take all practicable steps to mitigate the risk of ignition sources by allowing vehicles and ignition sources inbye the main fan in the restricted zone. Any one of those vehicles could have been an ignition source on 19

[38] Diesel-powered vehicles have the potential to provide a source of ignition for methane. This can occur if the vehicle operates in the explosive methane atmosphere (5 % to 15 %) and built-in protection systems fail (if a vehicle has a protection

system). Operation protection systems are designed to control the overheating of engine parts and exhaust systems and to ensure the isolation of electrical components from explosive atmosphere. These protective mechanisms will normally operate when the level of methane in the air rises to 1 %.

[39] Diesel motors may become supercharged when in a methane enriched atmosphere leading to out of control revving and engine overheating. This can provide an ignition source.

[40] There were a number of different types of diesel vehicles that were operated by the company in the restricted zone. The load haul dumper did have a methane sensor which would automatically shut the engine down if it detected methane in excess of 1.25 %. The specialised mining vehicle; Brumbies and Drift Runners which were operated in the restricted zone did not have this type of sensor installed on them. The SMV vehicle did have a valve on it, but this needed to be activated by the operator as an emergency stop in the presence of methane. Unless the operator had a personal methane detector, the operator would not know the presence of potentially explosive levels of methane.

[41] The cost of fitting these vehicles with methane sensors was a relatively small cost against the overall cost of the vehicles. In relation to the Drift Runner, it was

[42] It is known that at the time of the first explosion, a Drift Runner was operating in the B heading and well into the restricted zone. There was also the VLI drilling rig operating well inbye the main panel which had a gas sensor fitted but was faulty.

[43] All of the vehicles being operated in the restricted zone should have had methane sensors which automatically shut them down once the methane reached the level of 1.25 %.

[44] It is clear from the evidence that I heard, that PRCL failed to undertake a systematic programme of pre-drainage prior to the extraction of coal from panel 1,

[45] The purpose of the pre-drainage of a coal seam is to catch a gas at high purity from its source before it can enter the mine airways. Pre-drainage of coal reduces the in situ gas content of the seam and, therefore, reduces the level of methane emitted. It also reduces the likelihood of outbursts. Outburst is where the coal face very quickly releases a large amount of methane into the atmosphere. Pre-drainage is consistent with the Health and Safety in Employment Act’s hierarchy of controls set out in sections 8 to 10 of the Act. This requires elimination as the first step in managing hazards in that pre-drainage will eliminate some methane from the seam before mining proper begins.

[46] The company was very aware of the importance of pre-drainage and had commissioned a number of reports. These were from external consultants and pre- drainage of the coal seam was recommended in 2000 and 2006. More recently in May 2010, a report commissioned by Drive Mining Property Ltd proposed a gas drainage schedule and design developments. There are a number of key criteria for gas drainage including flanking drainage holes to be drilled five metres to 10 metres from the rib line of both main development roadways in extraction panels as a minimum.

[47] Post the explosion, PCRL provided a plan of the geological bore holes drilled at the mine. Importantly, the plan shows that panel 1, 1W1R did have a flanking hole, however, only on one side of the intake road. There were no holes drilled through or over the pillar or on the western flank of the pillar. Gas sampling taken prior to the explosions indicated that the methane content was 8.29 m3/t, therefore, it was of significant importance that the coal which was emitting high levels of methane was drained prior to mining.

[48] The pre-drainage would have removed some methane from the coal before mining and so would have reduced the risk of methane accumulation and the demand on the ventilation capacity.

[49] I accept that it is unlikely that the rupture of this pipeline itself would have contributed enough gas to have provided the full fuel source for the main explosion underground. It could though have caused an ignition, fire or minor explosion at the location of the rupture at any time. Although a budget had been allocated to this work, the company was not prioritising this work as at 19 November 2010.

[50] Methane was drained from the various dual in-seam bore holes by a drainage pipe line. The pipe line ran down the C heading through cross cut 2 pit bottom north to spaghetti junction at the intersection of C heading and cross cut 2, along C heading to fresh air base (FAB2) stub and along the roof of the FAB2. At FAB2 it went through a riser to the surface. At the surface, the methane rise was vented through a flame arrester to the above ground atmosphere; however, this was adjacent to the intake for the slim line shaft and also the FAB2.

[51] Of the extensive pipe line, 67 metres of that pipe line ran through the man and materials roadway. This included some of the most congested areas of the mine and particularly round a number of potential ignition sources, particularly at spaghetti junction.

[52] The location of the methane drainage range in the ventilation intake road also meant that in the event of leakage from the pipe line, methane would very quickly enter into the intake air before any ventilation control devices were closed.

[53] The company had given consideration to moving the drainage pipe line and had considered options for upgrading and relocation of the methane drainage range in 2010.

[54] In August 2010, they received a concept design report on methane gas drainage and utilisation from Mechanical Technology Limited.

[55] A former employee of the company, Mr Van Rooyen, who was the technical services manager up until the beginning of November 2010, told the informant that a budget had been allocated to upgrade the gas drainage range and a site had been selected for a new riser. The new site would have allowed for the relocation of the range from the intake roadway and man and materials route.

[56] Locating the methane drainage range in an intake airway and a man and materials roadway are well understood risks in the coal mining industry.

[57] As already indicated, one of the significant hazards of mining underground is the ventilation of the mine and the monitoring of the various gases which are inherent in the mining of coal, particularly methane and carbon monoxide which both at certain levels are both highly explosive.

[58] Real time gas sensors provide continuous readings of gas levels. There are also other devices such as tube bundling devices that are a very important backup tool to real time gas sensors.

[59] There were nine fixed real time gas sensors designed to detect methane installed at Pike River Mine on 19 November. These were grouped into three broad categories based on their purpose and reporting configuration. Five of those nine sensors were installed and positioned to de-energise the electrical equipment in the non-restricted zone that was not explosion protected or intrinsically safe. These detectors were designed to trip when the methane reached 0.25 % of the general volume of air. These sensors were continuously reporting to the control room and the recording system called SCADA. These sensors were in the non-restricted zone and outbye of the main fans. The second category of methane sensors were three sensors in the return roads which reported to the control room and the SCADA system. Of all

of these sensors, these were the most important given their location which was inbye the main fan and closer to the panel 1, 1W1R.

[60] The purpose of these sensors was to provide real time data for an immediate response to rising or hazardous methane levels. This would have enabled interrogation of the system to reveal the origin of the methane and its occurrence. On

19 November, only one of these three sensors was operational. The sensor that was closest to the panel on the return road had been out of operation since 13 October

2010 and the other return sensor at the base of the ventilation shaft had not been operational since 4 September. The company was operating outside of its own ventilation management plan (VMP). The plan required that electrical supervisors take all practicable steps to restore any failure in the gas monitoring system to fulfil working orders and that any delay be communicated to the mine manager and the ventilation engineer (as of 19 November there was no ventilation engineer). The VMP also required that a risk assessment be carried out and permit to work be issued before any mining activities could be carried out if any part of the gas monitoring system was not fully operational.

[61] There was no record of a permit to work or any risk assessment following the failure of these sensors. Post explosion, the mine manager, who by default was acting as the ventilation engineer, was unaware of the operational status of these sensors. The only return methane sensor that was operational was located at the top of the ventilation shaft, therefore, well removed from any potential source of the methane within the mine. It was also at a point where it was at maximum dilution in relation to the ventilation systems. Its location and placement at the top of the ventilation shaft was questionable. It was suspended within the shaft by a two metre long rope and exposed to both mud and potential impact on the side of the shaft, either of which could have affected the sensor’s accuracy. The sensor had shown evidence of defective readings around 5 and 6 October. The sensor had last been checked or calibrated on 4 November 2010. The calibration however, did not test above 2.5 % of methane and did not identify or address the earlier failings of the sensor. There was one other sensor located; however, this was attached to the hydro- monitor. It only reported methane readings to the hydro-monitor operators only. It was located in panel 1, 1W1R return road. There was no data recorded from this

sensor and therefore, it could not be trended or reviewed. Importantly, this sensor did not de-energise the hydro-monitor or any other electrical equipment in the nearby vicinity.

[62] Importantly, none of the sensors in the mine were capable of accurately measuring levels of methane over 5 %, this being the critical explosive levels for methane per general body of air. This meant that PRCL had no information as to how far above 5 % its readings of methane had been and could not calculate for example, the total amount of methane quantities once readings were available. The company did provide handheld protectors for use underground. These were used by deputies to check that methane remained at safe levels throughout the mine. The difficulty, however, is that methane is highly buoyant and it tends to lie near the roof. In most places of the mine, the roof was at least 3.6 metres high. The company did not provide its deputies with extension probe pumps so that roof readings could be undertaken.

[63] There were also real time sensors installed in the mine to detect carbon monoxide. This is important as carbon monoxide is an early indicator of spontaneous combustion which is a potential ignition source for methane. The hydro-mining at panel 1, 1W1R was one place in the mine at particular risk of spontaneous combustion because of the coal that was left behind after extraction. There was a real time sensor installed in panel 1, 1W1R return road. However, on 19 November this sensor was not operating and had not been operational since 13 October. Since that date, there had been no continuous readings of carbon monoxide made in the return road from the panel.

[64] Although there were personal sensors available to measure carbon monoxide, these sensors were unable to detect carbon monoxide below five parts per million. Therefore, early indications of carbon monoxide would not necessarily be detected by the sensors. The mine manager, Mr White, was unaware of this limitation. The other carbon monoxide sensor was at the base of the ventilation shaft at the point where all return ventilation air from the mine was combined. This was woefully inadequate. It was not able to provide any information to indicate what activity had produced the carbon monoxide or to determine the original location. It was also

unable to provide any information on the concentration of carbon monoxide at the point of origin.

[65] In addition, all of these sensors needed to be calibrated to remain reliable. The Australian standard maintains that there should be daily inspections and monthly calibrations. This was also the maintenance and calibration programme which was set out in the company’s VMP. As part of the investigation, the informant was able to access the MEX system, which is a system which computer generates work orders. These are tasked to various trades people or personnel within the mine to do. Once the task was completed, the tradesperson confirmed it had been completed and this was then entered back into the system by an employee. A review of the master MEX records, and all of the available work orders, show that in three months prior to the explosions, two sensors was definitely calibrated and at best, two other sensors were possibly calibrated.

[66] As there were at least five non-restricted real time monitors in place underground, and one at the top of the vent shaft, there should have been at least 18 calibration events (one for each sensor) over the three months.

[67] After hearing the evidence, I am satisfied that the company failed in a significant way by having inadequate real time monitoring equipment underground. Also, that this equipment was not maintained and calibrated and, therefore, significantly increased the risk for those persons working within the mine.

[68] With reference to all of the practicable steps, I see this as a major failing on behalf of the company and one that contributed significantly to the events of

[69] I have reached that finding on the basis of a number of events that were recorded, although not collated in any sensible way.

4 September - the real time sensor at the base of the ventilation shaft is non-operational.

22 September – extraction is commenced with the hydro-mining at panel 1, 1W1R.

13 October – the real time sensor in panel 1, 1W1R is non- operational, both for methane and carbon monoxide.

30 October – the pressure event which causes a VCD to be knocked over. Therefore, compromising the ventilation of the mine.

13 -16 November – over this weekend period, there was a significant panel shift with no risk assessment undertaken.

14 November – a gas spike in excess of 5 %. 15 November – a gas spike in excess of 5 %. 16 November – a gas spike in excess of 5 %.

17 November – a gas spike at 1.96 %, which lasted for over 45 minutes and was consistent with methane in the return headings of

19 November – the gas sensor on the VLI drilling rig inbye of the main panel was found to be faulty. The electrical circuits also had not been certified since June 2010. The rig was allowed to continue drilling.

Between the 25 October and 19 November, there were 13 occasions when the gas spike above 1.25 % of methane per normal volume of

Practicable step 5 – Inadequacies in relation to identifying reporting and investigating and resolving causes of gas plugs

[70] Everybody was aware that this mine had been a designated a “gassy mine”

[71] Therefore, there was a potential for explosive mixtures of methane to accumulate underground and to travel as plugs of gas through the mine’s ventilation system. The data retrieved from the company show that between 25 October and

19 November 2010 the sensor at the top of the ventilation shaft showed that on 13 occasions there were spikes in excess of 1.25 % of methane as recorded. It must be remembered, that this was at the most diluted point in the ventilation circuit for the return air to exit the mine. Of those 13, two can be attributable to the calibration of a sensor and the other two the restart of the main fan.

[72] Section 2 of the Health and Safety Act defines “accident” as an event that causes any person to be harmed or might have caused any person to be harmed. Accumulations of methane in the explosive range meet those criteria.

[73] The expert evidence I heard was that a methane spike measured in the ventilation shaft sensor that came from only one particular place or panel in the mine and measured 1.25 % was an “accident” event. This meant that the methane concentration at the origin inbye would be excess of 5 %. This is in the lower explosive limit for methane. This level of methane could be potentially higher if the original spike originated at panel 1, 1W1R.

[74] The last significant spike occurred on 17 November, when a peak level of just less than 1.9 % was recorded. This peak lasted for over 45 minutes and it is estimated that within that time 1,200m2 of methane exited through the ventilation shaft. This coincided with the hydro-monitor extraction report where the crew were degassing the panel. The quantities of methane in the return suggest it was running close to 7 % which is in the highly explosive range for a period of at least 10 minutes. Although this data was recorded, this was not reviewed as an incident report. There was only one occasion (12 October) when an accumulation of methane

was recorded as an “incident”. This was methane found in the monitor housing room which was in the non-restricted zone.

[75] There is a well recognised need to have a procedure in place to manage gas plugs or spikes. It is recorded in the Guidelines for Ventilation of Underground Mines and Tunnels 2009. This was also recognised by the company’s own Operational Preparedness Gap Analysis (OPGA) which was prepared in August

2010. Although the control room at Pike River monitored the gas alarms, the control room operators were not required to record details of the gas spikes or gas alarms. Of

the 13 occasions when gas was recorded at over 1.25 %, only five of these were recorded in an event book. The lack of monitoring and reporting was evident when the mine manager, Mr White, was interviewed. He was unaware of gas spikes in excess of 1 % after the new fan had been installed and commissioned on 27 October

2010. The company was well aware of its obligations as it had initiated a process to record and investigate methane spikes. There was to be a control room gas alarm log that was designed so that gas alarms would be acknowledged and notified and investigated. However, as of 19 November, there was no methane alarm recording and no investigation systems were in place.

[76] A tube bundle is a system that draws samples of underground atmosphere up to a tube by suction to the surface for analysis and then reports those results to the analyst to the control room. The benefit of a tube bundling system is that it can continue to work even if there is no power underground, such as that which may occur in an explosion.

[77] Tube bundling is not a real time monitoring because depending on the distance the atmosphere being tested may take five to 10 minutes to travel to the analyser. However, it does allow for continuous monitoring over time which means that trends can be identified, analysed and if necessary responded to.

[78] There was an intention by Pike River Coal to install a tube bundle system. Various reports were commissioned in June 2006 which recommended a tube bundle system. The company’s own ventilation management plan also makes extensive reference to tube bundles and a reference to collection of sample bags at locations specified by ventilation officers. As late as 1 June 2010, there is reference to a budget for the purchase of a tube bundle system from Australia.

[79] A Deputy report on 3 November 2010, (17 days before the explosion) discussed the organisation of regular bag sampling, both at the panel and with the hydro-co-ordinator. However, no action was taken despite that matter being raised. On 19 November there was no tube bundle system in place.

[80] In my opinion, having heard all of the evidence, I am satisfied that the company failed to instigate what was a relatively simple and cost effective means of sampling the gas in the mine. This sampling would have provided the extra assurance that the real time system was operating accurately and that there was no sign of spontaneous combustion underground. This may well have assisted post- explosion in relation to the information required to be gathered to enable retrieval of the miners.

[81] The effect of a significant roof fall in the goaf area or a windblast event causes large quantities of methane to be pushed into the mine’s workings. Windblast is generally accepted to be a goaf roof fall leading to air movement with the velocity in excess of 20 metres per second. An air pressure detection system is a series of pressure transducers, paddle switches and micro switches that are installed so that when an overpressure wave from a wind blast or goaf fall is detected, any electrical equipment is switched off. There was no pressure transducer or micro switch installed, either in the return or the intake road at the panel 1W1R on 19 November. This meant in the event of a pressure wave being pushed down the return or intake road, that there was no mechanism to limit the potential electrical ignition sources inbye.

[82] These could be relatively simple devices which would not be expensive to install and on the evidence I heard were particularly important in the panel intake and return headings.

[83] An outburst event is a well recognised occurrence when one mines coal. This is a particularly known event in West Coast mines. In 1997 two West Coast miners were killed at Mount Davy Mine when an outburst released an estimated 12,400m2 of methane. Outburst events cannot be easily predicted. However, conditions where they can occur are reasonably well known. One way in which this hazard can be managed is to determine the propensity of the seam which is being mined to an

outburst event. Once this is known, appropriate control measures can be put in place such as pre-drainage. The New Zealand Minex Code of Practice for underground mining and tunnelling states:

Potential for outburst/rockburst must be quantified during exploration phase and an appropriate risk assessment completed including independent review.

[84] Prior to 19 November 2010, there were a number of external consultants who gave advice to Pike River Coal, recommending that they examine and quantify the outburst potential for the seam. These reports were as early as 2000 and recent as September 2010. The defendant company also appeared to have recognised the need to quantify their outburst potential. In 2009, an outburst management plan was issued, although as of 19 November 2010 it was still in its draft form. The draft plan stated “the basic principle of operation was that no mining will take place when the gas content of the coal is above the established outburst threshold level.” As of 19

November, no outburst threshold level had been established or stipulated. The risk of outburst was also noted in the formal risk assessment of ventilation and gas monitoring which was undertaken prior to the hydro-mining, which commenced on 7

[85] Despite receiving numerous reports and recommendations and the company recognising this hazard, the company did not establish the potential for outburst so that it could then take a systematic approach to managing it. To commence mining and extraction prior to knowing the outburst potential of the coal seam was, in my view, extremely dangerous. Although it is unlikely that the outburst was the source of the methane from the explosion on 19 November, it cannot be excluded as a possibility.

Practicable step 9 – Machines without fully operational calibrated methane sensors

[86] This step relates specifically to the VLI drill rig which was working inbye of the main panel on 19 November. On the morning of 19 November, a calibration check detected a fault with a methane sensor. The fault was noted and a new work order was generated to replace the sensor. However, the drill rig continued to

operate. It was also noted that the sensors and ATA calibration had expired in 2010. The ATA calibration is a third party verification undertaken every six months to confirm that the gas sensors are operating as designed.

[87] There was no formal systemic risk assessment carried out to determine if it was safe to continue to operate the drill rig without the calibration. This was directly in contravention of Pike River Coal’s own ventilation management plan that states:

No road header or continuous miner should operate without a fully functional MMGM (machine mounted gas monitoring) system. Any failure of the MMGM shall be reported to the production deputy who shall arrange the repair or replacement of the failed equipment and tag the machine out.

[88] There was no written procedure by the company to prevent the use of any machine or to require a formal risk assessment to be undertaken if the sensor was faulty.

[89] Gas detectors on these machines are a fundamental control to prevent electrical equipment being a potential source of ignition. This allows the machine to shut down on its own without recourse to the driver of the machine being aware of methane at a critical level.

[90] This was a critical failure by Pike River Coal in relation to ensuring the safety of any person entering the mine.

Practicable step 10 – Identification and investigation methane drainage inflow and output rates

[91] Investigation has shown that on 19 November the difference between the quantities of gas measured entering the line and the outflow of gas measured exiting the riser was 113.2 litres per second. The discrepancy means that there was significant amounts of almost pure methane which was unaccounted for in the underground mine.

[92] The failure to detect or investigate the discrepancy meant that the company could not ascertain whether the methane that was unaccounted for was leaking from the pipeline into the mine or was back-feeding through the bore holes in the goaf or

other headings or whether there was a fundamental fault in the measuring equipment. Having said this, however, the unaccounted for methane was not in the range sufficient to fuel the explosion on 19 November. The effect of having methane being released from leaks in the system or back-feeding through the other bore holes would have increased the methane load on the overall ventilation system and may have led to added layering in the areas of low ventilation, which were those areas significantly inbye of the panel.

[93] I am satisfied that this was a simple and practicable step which was available to a company which they failed to take.

Practicable step 11 – Risk assessment for the secondary extraction panel

[94] Secondary extraction occurs when a mine begins to develop an extraction panel from the development roadways. A shift from primary development to secondary extraction began at the Pike River when the mine began retreat mining in panel 1, 1W1R on or about 22 September 2010. The extraction process itself involves a change in mining methods, practices and hazards. As such a second risk assessment was essential to identify and assess and develop controls for these risks. The failure to address the hazards and risks associated with secondary extraction in a holistic manner means that hazards are likely to be approached on an ad hoc basis without considering the interaction and aggregation of these risks.

[95] The practice of carrying out second working risk assessment is well established in Australia and is codified by the Queensland Coal Mining Safety and Regulations 2001, division 3 Regulation 317. Mr Reece during the course of his evidence explained to me that no extraction can take place without a risk assessment having been undertaken and this risk assessment is then approved by the inspectors of the mines. There are certain statutory requirements that need to be considered in relation to the risk assessment which relate to the geology, the gas release, pillar stability, method and sequence of coal extraction, along with methods relating to the support and control of the strata ventilation, controlling of spontaneous combustion events and methods necessary to achieve stability within the goaf.

[96] The commencement of the extraction plate process at the panel meant that there was a considerable amount more methane available. This meant that an equivalent square area of goaf in the panel had the potential to contain more than double the overall volume of methane compared with a long wall goaf. This is due to the height of the seam at Pike River being 10 to 12 metres. This effectively formed a reservoir and allowed a large amount of methane to be available for a potential explosion given the right combination of oxygen and ignition source. Secondly, due to the nature of the extraction and an extension of the area that was to be mined, this increased the risk of a cave collapse or a large plate caving off the roof. This could lead to an overpressure event which would push methane down the return roads and into the ventilation workings of the mine. Further, the design of the panel provided for two relatively short headings which went directly onto the main return. The main return was isolated from the main intake by single temporary stopping at crosscut 3,

1 west. This was directly in line with the return panel heading and any overpressure event could be expected to impact directly on this stopping.

[97] The risk associated with such straight and relatively short returns was that any overpressure from a roof fall could not be dissipated by multiple headings or by the length of the headings. The other risk was if the stopping at crosscut 3, 1 west failed then this would introduce methane into the intake and carry the diluted and potentially explosive mix into the inbye workings where there were a number of ignition sources present. As will be seen later on in this decision, there was also inadequate ventilation inbye of the main panel.

[98] On 19 November there were at least three working places inbye of the main panel, each containing a number of items of electrical equipment that could be potential ignition sources.

[99] The company was on notice of the risks. There were several reports by external consultants who had identified the risks of spontaneous combustion in the goaf. They had also developed a spontaneous combustion management plan. However, this was still in draft form as of 19 November 2010. The lack of a risk assessment also overlooked the risk of a frictional ignition. Frictional ignition is when items fall onto the floor of the mine and cause a spark. In 2008 an incident had

occurred whilst a road header was being used to excavate a floor. Subsequent investigations into the lithology of rock revealed that the seam floor had high potential for frictional ignitions. Further study, as late as 2010; found that the rock in the strata between the coal seam and the rider seam was potentially incendive. This was also true of the rock from a bore hole above panel 1, 1W1R. A plan was implemented in 2009 to deal with the frictional ignition issue; however, the plan was incomplete. It was still in draft form as of 19 November.

[100] Another assessment taken by the company in September 2010, the windblast risk assessment, considered the risk of frictional ignition. There was reference to using water as the mining method. However, extraction and the use of water were not continuous in the panel.

[101] The SCADA data and electricity usage indicates that no mining took place three hours prior to the explosion on 19 November. And similarly, on 17 November, prior to the gas spike discussed above, there was no working for 10 hours. The company had conducted a number of risk assessments; however, these were on an ad hoc basis and in my view, perfunctory and inadequate.

[102] For example, the windblast assessment undertaken by Pike River Coal did not quantify or model the potential windblast effects or quantify the distance a pressure wave could potentially travel into the area of the mine. It also appeared incomplete. There were no due dates, completion dates or sign offs for completion. This risk assessment proposed a number of controls such as dilution doors, rated ventilation structures, sacrificial stopping wooden doors in crosscut 3, 1 west (an important stopping), a degassing procedure, updated emergency response plan, and roof and floor coring. None of this was implemented as of 19 November.

[103] Some of these suggestions were also mirrored in the operational preparedness gas analysis prepared on 26 August 2010. This indicated critical safety systems including dilution doors, protection and goaf ceiling as needing to be identified, checked and implemented. Other actions - gas plug procedure, panel ventilation, fan and ventilation trigger action response plans, planned evacuation and fire fighting plan – had not been fully actioned or implemented as of 19 November. Of real

concern is that from 13 October there was no real time monitoring of gas readings in the panel return road. Therefore, there was no continuous monitoring for the presence of methane or carbon monoxide. If there had been recurrent methane spikes or plugs from this panel, this should have prompted a reassessment of the adequacy of the methane control in the panel and return roads. Prior to 13 October when the sensor ceased to operate, methane in excess of 5 % was present in the panel return road on numerous occasions. Although dilution doors had been mentioned in a number of reports, as at 19 November, the dilution doors were not operational.

[104] An incidence had occurred on 30 October when a large roof fall had occurred in the goaf, which knocked over the stopping in crosscut 1 panel 1, 1W1R. This should have triggered a reassessment of the adequacy of all of the stopping that were in the vicinity of the panel, in particular the stopping directly opposite the return road.

[105] I am satisfied that the company did fail to carry out the second working risk assessment and this was a practicable step that should have been identified and systematically managed. This would have alerted the company to the hazards and risks associated with the work in the panel. Although there were individual risk assessments taken, they failed to identify a number of risks and failed to implement them or were not implemented at all.

[106] PRCL engaged Valley Long Wall Drilling Pty Ltd to carry out inseam drilling at the mine. As a result a track-mounted Boart Long Year LMC75 direction rig was used for drilling. Although it was operated by VLID, PRCL as part of its contract agreed to carry out statutory electrical inspections on the drill rig and to provide up to five hours of assistance a week were there any electrical problems. Although the electrical inspections were weekly and weekly gas sensor calibrations, the records from Pike River Coal record that the last electrical inspection carried out on the rig was on 29 June 2010. Although there were work orders generated between 29 June and 19 November, none of these work orders were recorded as completed; they were recorded as not done.

[107] The risk of faulty electrical equipment igniting methane is a well recognised underground mining issue. Although the company intended to manage this hazard through preventative maintenance, this simply was not carried out. In respect to this particular piece of equipment, the scheduled maintenance had not been implemented for four and a half months prior to the explosions. More importantly, this was a drill rig which was being used to operate at a face and was exposed to methane from the bore holes it was drilling.

27 October 2010 was the main fan which operated the ventilation at the mine. This fan was located underground. The motor was in the non-restricted zone of the shaft and the driving motor was connected to the fan through a concrete bulkhead. The fan was located in the return air in the restricted zone and, therefore, had the potential to be a source of ignition if the percentage of methane in the return air was above 5 %. Therefore, there was a necessity for regular mechanical inspections. The manufacturer’s instructions recommended a minimum of a daily visual inspection and a check on the bearing temperatures and vibrations levels. Again, the schedule of work and inspections were generated by the MEX system and a mechanical maintenance checklist attached, which was to be completed by the fitter. For the 18 days prior to the explosion, 27 work orders were generated that required an inspection of the main fan. Four of those were fully completed, 14 were not done and nine were only partially completed and did not involve official inspection of the fan. Importantly, from 1 November to 18 November, the tag recordings which record the temperature in the C motor, recorded a fault.

[109] I am satisfied that the company failed to take all practicable steps to ensure that they complied with an inspection regime in relation to this very crucial piece of equipment. Also, that they should have exercised sufficient oversight of the inspections that were carried out to ensure that the inspections were thorough and in accordance with its own procedures and the manufacturer’s recommendations.

[110] To commence mining, the company required a permit. The first permit issued to the mine for panel 1, 1W1R was dated 22 September 2010. This allowed extraction involving a total extraction width of approximately 30.4 metres. This was made up of the pillar of 20 metres and a 5.2 metre intake and return heading on either side of the pillar. The company anticipated that the roof of the goaf would progressively collapse up to the seam as the extraction in panel 1, 1W1R progressed. There was an expectation that the massive sandstone roof would act as a laminated beam.

[111] As discussed, the main risk associated with a goaf roof fall is windblast. Windblast is a roof caving that produces air velocity in excess of 20 metres per second and would typically occur when the roof fails as a plate over a sizeable area. The most severe consequence is a gas or coal dust explosion. Other hazards are displacement of people and equipment.

[112] The company had received expert advice that the roof strata in the panel had geological characteristics that are associated with windblast. It is accepted, however, that on 19 November the roof area of the goaf was not large enough to produce a windblast effect.

[113] However, a large roof fall can produce similar outcomes to windblast in terms of expelling methane into the workings of the mine but with lower velocity winds. Although you might not get the same effects of wind velocity, a large roof fall will still risk gas being expelled down the return and intake roads. This scenario was considered the most likely explanation for the sequence of events that led to the explosion on 19 November. As noted, the company was warned of the risks of windblast occurring in this panel in the goaf. A report by Strata Engineering Property Ltd on 29 August 2010, noted:

This places an emphasis on the ongoing collection of structure data i.e. mapping and core logging to access the structural environment on, initially at least, a panel by panel basis.

[114] The mine was granted a second permit on 15 October 2010 to increase the extraction limit for the panel by 15 metres on the east and right-hand side of the intake road to a total of 45 metres to allow recovery of more coal. This moved the extraction closer to a geological fault that flanked the panels. The faults may act as release plains for roof strata if they create a break in the continuity of the supporting roof strata. The increase to 45 metres is represented as a 50 % increase in the total panel extraction width. On 25 October Pike River Coal was provided further advice from Dr William Laurence from GeoWork Engineering Property Limited. That advice indicated that although minimal caving was indicated for the 30 metre wide panel, an increase in the risk of caving was indicated for the 45 metre wide panel. It was also noted that extending the panel had decreased the strata stability against the flanking normal fault. A letter also clarified that “due to lack of data, critical perimeters had been assumed, which does result in some uncertainty”. The company was aware of the uncertainty and therefore, attempted to take further samples from the intake road with the use of a highlander drill. This was attempted in the first few weeks of September; however, the highlander drill could not be made operational due to a lack of air pressure.

[115] The operations report to the Board of 13 September 2010, shows further that the Board was advised:

The roof conditions have been evaluated by bore scoping but core drilling would be valuable to confirm the results and give comfort in terms of the recommended and installed roof support in the panel. The fact that this information has not been gathered does increase the risk in this panel.

[116] Despite the earlier problems with the highlander drill, the evidence retrieved from the company note that on 21 October, this was able to be successfully started. However, no further core samples were taken prior to 19 November.

[117] The extraction of coal from panel 1, 1W1R was commenced with very limited geological knowledge and with the prior knowledge that the conditions associated with windblast were present in the roof strata. Therefore, any work should have been approached with extreme caution. The fact that they increased the extraction width by 50 % dramatically increased the risk of plate fall or roof caving. Although there was an endeavour to collect further information, this was not

implemented and was not given priority. The company, however, went ahead to increase the extraction width without any further knowledge or investigation.

[118] I am satisfied that was a crucial and fundamental error on behalf of the company and a fundamental breach of its obligations to those persons entering the mine. The company should have delayed the increase in extraction width of the panel until it had obtained the further information with respect to the possible caving.

Ventilation management – CRN 204 and 208, designing, building and maintenance of a ventilation control device (VCD)

[119] I find that the Defendant failed to take all four practicable steps as set out in the amended particulars.

1. PRCL should have designed, built and maintained adequate ventilation and control devices.

[120] VCDs are a very important aspect of the ventilation of a mine. They control the quantity of air that is proportioned to the different sections of the mine and separate the contaminated return air from the fresh intake air. Stoppings are the most common type of VCD. They are designed to be full partitions that isolate the differential pressure and resulting potential air flow between the intake and return loads. The stopping may have a door built into it to allow access for personnel and all vehicles. Stoppings are predominately located in the crosscuts between the headings.

[121] Inadequate or damaged stoppings will leak air from the intake to the return and reduce the amount of air available to sections inbye the leak. Major damage or loss of a stopping will lead to a short circuit of the ventilation circuit and a significant loss of inbye ventilation.

[122] Failure of a stopping as a result of an overpressure wave from a roof fall may result in the increase of methane gas into the intake airways. As the intake airways supply fresh air, concentrated methane in the intake will become diluted and may pass into and eventually through the explosive range. Also, the intake air always

provides air to the working faces inbye. It will therefore take any potentially contaminated air into working faces where there are multiple sources of ignition.

[123] Stoppings can either be permanent or temporary. Although temporary stoppings are required as the mine is developed, because they are less robust and will tend to leak more air, permanent structures should be in place as soon as practicable to do so. PRCL did have a standard for stoppings. This was set out in its underground standard SOP V1. This design standard was inadequate. Its design for permanent and temporary stoppings was the same. There was no evidence to suggest that the company had an engineered design for permanent stoppings between the mine roadways. The only difference in practice between a temporary and a permanent stopping was the permanent stopping had mesh over a wooden and brattice sub straight and was sprayed with a layer of cement-type product.

[124] During the course of his evidence, Mr Reece discussed the inadequacies of such a temporary type structure and in contrast highlighted those permanent stoppings which are used and are a statutory requirement for mines in Queensland. In Queensland, stoppings are pressure rated and designed to withstand an overpressure of 35 kpa. These stoppings are usually made of solid concrete and steel and are significant structures.

[125] If the Australian Standards had been applied to the Pike River Mine, all stoppings between the main A, B and C headings would have been built to withstand an overpressure of 35 kpa. This included the essential stopping in place at the cross cut 3.1, W which is opposite the end of the return road for the panel. The experts consider that the most likely scenario for the course of events leading to the explosion on 19 November was the failure of this stopping which led to methane- enriched gas being expelled into the intake airways and being exposed to ignition source inbye. It has been estimated by the experts that the overpressure from a roof fall in the goaf would have been approximately 10 kpa. A stopping, which was rated up to 35 kpa or of substantial construction, would have withstood the overpressure. This would have meant that the gas expelled would have gone down the air return road and not mixed with the fresh air.

[126] Importantly, and in my view telling, was that the stoppings that were in place at cross cut 3 and 4,1 W were not constructed in accordance with the company’s own underground standards, inadequate as they were. They were variously described by witnesses as pogo sticks or wooden posts with brattice cloth attached.

[127] Such a design has no ratings given to them by the manufacturer. These stoppings were not consistent with the company’s own ventilation management plan which contained a clause:

No secondary development panels may be turned off from a main development panel unless the ventilation stoppings in the main development panel are of a permanent design inbye the turn off position.

If this had been complied with, this would have included the installation of permanent stoppings in cross cut 3 and 4 before the panel was commissioned.

[128] The company had plenty of time to create some permanency for the stopping at cross cut 3. On the ventilation survey dated 2 September 2010, it is shown to be in place. Therefore, the company had at least two and a half months to strengthen and rate this stopping prior to the explosion.

[129] The stopping at cross cut 4,1W was first shown in place on the ventilation survey dated 9 November. This stopping was being worked on 19 November, however, whether work had been completed or not by the time of the explosion is unknown. The company was aware of the vulnerability of the stopping at cross cut

3,1W. In a windblast risk assessment dated 3 September 2010, the document noted that additional control was needed to manage the risk.

[130] Seven weeks prior to the explosion, the effective overpressure from a roof fall on a stopping had been demonstrated. A number of documents supplied to the informant described a roof cave in the goaf at approximately 4 am on 30 October that knocked over the entire stopping at cross cut 1 between the return and intake road to panel 1, 1W1R. This stopping was propped and wedged in place but subsequently rebuilt and bolted into the roof.

[131] The stopping in cross cut 2 ABN, was in the direct line with the return and served to isolate the main fan motor which was not explosion protected from the main return air. This stopping was constructed of mesh and a sprayed cement product over a wooden frame. The man door in it was a brattice cloth flap weighted along the bottom. This would only seal if the pressure differential between the intake and the return held in place. If there had been an explosion, this would have affected the pressure and, therefore, compromised the seal. Mr Reece during the course of his evidence was scathing of the woefulness of this stopping and in all of his experience in Australia had never seen a stopping of this nature constructed.

[132] VCDs are essential in maintaining the ventilation of a mine. Although VCDs located close to the explosion centre may be destroyed, those further away and outbye the explosion would have more chance of withstanding the effects of an explosion overpressure and remaining operable. If the VCDs remain operable, the ventilation system has a chance to dilute and reduce toxic oxygen, displacing fumes from parts of the intake airway. An operable ventilation system will control the ongoing accumulation of methane that may fuel further explosions. This enhances the chance of survival for any person underground and reduces the risks associated with re-entry.

[133] I am satisfied that the company failed to take all practicable steps to ensure that it had adequate VCD devices in place. I see this failing as being a significant contributor to the effects and subsequent explosions that occurred on 19 November

[134] If stoppings had been built to a higher standard or had been of a substantial construction, it is likely that any methane expelled from the roof fall would have been contained in the return and therefore, would have been exposed to fewer potential ignition sources. Furthermore, it is possible that adequate stoppings outbye the explosion may have remained operable following the explosion and enhanced the ability of re-entry to the mine.

2; PRCL should have ceased mining to allow for the risk of unpredictable but foreseeable events

[135] An effective ventilation system will supply enough fresh air to reduce methane concentration well below the lower explosive limit as quickly as possible. Methane is constantly released from the coal underground. Most methane is released when coal is first cut and emissions will reduce over time. The ventilation system progressively mixes and dilutes the methane to below the explosive range. The air quantity (metres per second) must be sufficient to safely dilute the methane and the air velocity (metres per second) must be sufficient to mix the methane and avoid layering or stratifying of the methane on the roof of the roadways and workings. The purpose of the ventilation circuit is designed to ensure that contaminated air is removed from the mine without re-circulating through the working surfaces and putting it into contact with ignition sources.

[136] In October 2010, a ventilation consultant engaged by the company noted that the mine’s ventilation system was “almost already restricted from the purely quantitative prospective”. In other words, the ventilation system was extremely fragile, particularly inbye of the panel which was being worked on.

[137] The experts have concluded that the fan curve provided by the manufacturer indicated that there was a total mine ventilation quantity of 114m3/s available at the setting that the fan was running on 19 November. Given the number of areas that the circuit ventilation circuit was required to ventilate (being six working areas) the quantity required to ventilate that amount of space was between 150m3/s.

[138] The modelling shows that in the week prior to the explosion, the quantity of ventilation available underground inbye of the panel split was extremely fragile.

[139] All of these areas indicate that they were all over velocity general up dip areas, and therefore, likely to have methane layering in the roof. The modelling undertaken by the experts also indicated that a small change in mine resistance such as a vehicle parked in the roadway or changes to the auxiliary ventilation fans was enough to significantly disrupt the flow and lead to re-circulation.

[140] In addition, a full stopping was set up behind an auxiliary fan at 1W2R. This meant that the fan was situated in a dead end without fresh air flowing over and around it. Modelling undertaken by the expert’s show that the stopping was necessary to get the ventilation circuit to work at the inbye phase due to the lack of pressure and volume of air in the circuit. The location of the stopping and the placement of the fan were inconsistent with the accepted mining practice in Queensland. This stipulates that there be at least an excess of airflow provided by the main ventilation system in excess of 30 % of the capacity of the auxiliary fan. This helps to ensure that there is sufficient airflow over the fan motor to prevent the fan being exposed to re-circulating air and accumulating methane. The company also specified a minimum of one third (33 %) of the air volume flowing across the fan. The risk of the placement of full stoppings behind fans was brought to the attention of the company by its ventilation consultant in April 2010.

[141] In the three days leading up to 19 November, there had been unprecedented metres of roadway cut in the ABM heading and a geological bore hole intercepted. Both of these events increased the amount of methane released at the headings with the latter gassing out the ABM heading on a number of occasions.

[142] I am satisfied that the company failed to take all practicable steps to ensure that it had adequate ventilation over all workplaces. I am satisfied that it was trying to ventilate too many places in the mine on 19 November and that it had inadequate quantities and velocities of air inbye of the panel split. This meant that the ventilation inbye the panel was insufficient for safe ventilation arrangements. It also meant that there was no buffer or excess available to allow for unpredictable but foreseeable events such as roof falls or leakage from bore holes.

[143] The Minex guidelines for the ventilation of underground mines and tunnels states that:

Any major change to the ventilation system should be modelled prior to the change being implemented. This is so the modelling will confirm the effect

of the change and all ventilation splits in the site and that all relevant standards can be maintained.”

[144] Between 12 and 16 November, the company undertook a major change to the ventilation system when it constructed a new overcast at the end of section C heading 1 west and crosscut 5, 1 west. This was known as the panel move. Two auxiliary fans were also relocated.

[145] Prior to the panel move, there was no evidence of any modelling being undertaken. Modelling should have been undertaken and would have been able to calculate the quantities of air to be supplied to each face and where stoppings and fans would need to be placed. The company had engaged consultants to use computer ventilation programmes and had that available to them. This could all have been undertaken on site.

[146] The company were able to provide (post-explosion) a plan of the panel move that was drawn by the underground operations manager. However, this plan was not accurate. This plan showed a crosscut at 3,1WAB as having been completed and forming part of the ventilation circuit, when in fact, it had not been started. This meant that while the plan assumed that the road header was cutting in a short stub with approximately 70 metres of vent cans back to an auxiliary fan, the actual setup was the road header was cutting in a 70 metre long dead end heading with approximately 175 metres of vent cans back to an auxiliary fan. Vent cans are the tubes that convey the air extracted to the face by the auxiliary fan and back to the fan motor itself then to the point where it is discharged. A longer run of tubes will create more friction, more leakage and affect the efficiency of the auxiliary ventilation. In addition, a longer dead end heading will require more air for efficient ventilation.

[147] It also apparent that a number of the deputies were unsure or unclear over the settings for the fans in the construction of the temporary stoppings following the panel move. One deputy had asked “what is the ventilation plan?” Pre-modelling would have determined to a large extent the settings and locations of the fans and the type of stoppings required before the panel move was undertaken.

[148] The last ventilation survey taken by the company was 9 November, which was seven days before the panel move took place. There was no reassessment or validation of the ventilation model after the panel move on 16 November. Experts have concluded from modelling that as of 19 November inbye the panel air split, the ventilation quantities were insufficient for safe operation. A ventilation survey would have bought this to the attention of the company.

[149] I am satisfied the company failed to take this practicable step, that it had the means to do so, that this was not a difficult task to have undertaken, and they were well aware of the difficulties of the adequacy and quantity of ventilation inbye of the panel split.

[150] Adequate ventilation of a mine is a key step in maintaining the safety of those persons working within the mine or entering the mine. Because of the inherent nature and dangerousness of the mine and in particular of this mine, adequate oversight of the ventilation system was crucial. Adequate oversight would include:

Accurate knowledge of the quantity and velocity of air, both in the main circuit and auxiliary circuits.

Location and condition of the ventilation control devices and the status and results of the real time gas monitoring.

Setting expected performance standards and specifications, including the development of appropriate TARPS and minimal ventilation quantities.

Monitoring of the performance of the ventilation system through scheduled inspections and audit surveys.

Assigning roles and responsibilities to competent persons and ensuring those roles are adequately resourced and performed.

Ensuring that information is transferred from those who collect it to those who need to know it and where information is not being

[151] These detailed arrangements were reflected in a ventilation management plan that was finalised in November 2008. The VMP however, was deficient for a number of reasons. There are a number of key procedures and documents that were not developed, including an alarm log book, seals and stoppings log book, stoppings SOPs and TRPs for spontaneous combustion. There are various responsibilities set out in it that were assigned to roles such as a ventilation engineer that were non- existing positions at the mine as of 19 November. Some of these deficiencies were bought to the company’s attention in March 2010 when the plan was reviewed by a commissioning engineer. The review noted a significant number of issues that required follow up before the commencement of hydro-mining and a number of these issues remained outstanding as of 19 November, including the review of the ventilation management plan prior to extraction. The provision of a ventilation TARP and a SOP for permanent ventilation structures and spontaneous combustion and confirming what ventilation structures were required prior to the hydro-monitors starting up.

[152] The company was aware of other deficiencies in their ventilation management plan and requested a ventilation consultant, John Rowland, to revise the plan in September 2010. Although Mr Rowland had read the plan, he had not commenced the review as at 19 November 2010.

[153] The VMP itself required that a review would be performed before the start up of hydro-mining and thereafter on a two-monthly basis. There is simply no evidence that this was completed. The VMP supplied to the Department after the explosion is the version dated 18 November 2008.

[154] A key component of the system was the oversight of the ventilation system and in particular, the adequacy and location of the VCDs. The plan required:

Ventilation plan shall be drawn up to adequately reflect the ventilation system at the mine to show the location description of every ventilation device.

There is no evidence to indicate that the company had a centralised system to record the design construction maintenance condition and location of stoppings.

[155] Post the explosion, and in the course of the investigation, the company provided a number of plans to the Department. The information in these plans is inconsistent with information from interviews undertaken by the investigation team with those who were underground. The plans do not show all stoppings and regulators that were in place on 19 November and do not show the location and type of all the doors or regulators. They also do not record the construction of the VCDs, and fail to record the settings or size of openings or regulators in the mine.

[156] There is also a requirement in the Minex guidelines for ventilation devices to be inspected daily for leakage and the results entered into a specified book or log and that this log is checked on a monthly basis by the site manager. The company’s VMP also mirrors these requirements. The investigation however, unveiled that there was no stopping book or log or central record for the mine detailing these stoppings. Although the deputies were required to complete a deputy statutory report for each shift, none of these reports specifically required the deputies to check on the condition of the stoppings, although they are required to check such items as telephones and toilets. The investigation found that there was no record of the system to carry out a more thorough site-wide audit of conditions of all the VCDs, no evidence that a systematic site-wide audit of VCDs was being undertaken. Although there was no explicit requirement of deputies to note the conditions of the stoppings in the statutory reports, there were notations of difficulties with some of the stoppings. Although these reports were communicated to the operations manager on at least four or five occasions prior to 19 November, there is no record that any remediation of the defects had been undertaken prior to 19 November.

[157] At the time of the explosion, there was no designated ventilation manager. Mr White was the general manager of the mine and the role had been delegated to him by default as there was no appointment made for the position. The company had

intended to appoint a person to carry out at least some of the roles of the ventilation officer or ventilation assistant; however, this had not been initiated as at 19

[158] Adequate oversight of the ventilation system was essential to ensure the ongoing safety of those underground. Based on the evidence that I have heard, the failure of the company to undertake this practicable step to ensure and manage the ventilation system through a holistic approach placed all persons entering the mine in extreme danger and it was a contributing factor that led to the fatal explosion on

Practicable step 1- Adequacy and function of explosion protection for the surface infrastructure

[159] As discussed above, adequate ventilation is crucial in the safe working of a mine. It is also a key step towards recovering a mine after an explosion. The continuation of adequate ventilation enhances the chances of evacuation or rescue of any person who has survived; makes re-entry safer and also helps control any subsequent build up of methane and secondary explosions.

[160] The company had installed a standby surface fan to act as a back-up for ventilation if, for instance, the underground fan was damaged in an explosion. The underground fan did not continue to operate after the explosion on 19 November. The stand-by fan also failed to start after the explosion. It sustained substantial damage to its blades and by-pass louvers from the explosion due to overpressure and debris exiting the ventilation shaft. The discharge also knocked out the control panel. This was caused through overpressure exiting through the personnel access door. The falling panel then depressed the emergency stop button on one of the diesel generators, both of which needed to operate to start up the surface fan.

infrastructure. They are designed to allow overpressure to by-pass the fan itself

before any damage results. Flaps were installed on the standby fan at the top of the fan cowling.

[162] Although there is no specific guidance on the sizing of explosion venting, there is some guidance available from the US Code of Federal Regulations. These stipulate that the vent area must be greater or equal to the connection entries. In accordance with this regulation, the explosion flaps at the Pike River Mine should have been at least equal to the cross section or area of the ventilation shafts.

[163] The explosion flaps installed at Pike River were less than half the size they should have been. The ventilation shaft having a diameter of four metres, giving the total area of 12.56 metres2. The flaps in place on the cowling had a total venting area

commissioning engineer engaged by Pike River Coal for the main ventilation project noted in March 2010:

Needs further detail on provision for explosion protection and what, if any, implications this may have for the vent system operation and design.

Although this matter was raised, no further feedback was made by the company. Alternative arrangements had been suggested as early as 2000. These suggested having an additional excavation shaft running at an angle off the main shaft and connecting to the surface. This angled shaft continues to the ventilation fan and is the route for return air during normal operations. Large explosion doors are built over the top of the main shaft to seal it in the event of an explosion. Overpressure will tend to follow the easiest and straightest route, therefore, exit through the explosion doors on top of the main shaft, bypassing damage to the fan itself in an angled shaft.

[164] The US regulations also stipulate that the surface fan blades should be offset by a minimum of 4.6 metres from the nearest mine opening. At Pike River, they were located 2.3 metres from the ventilation shaft. This made them far more susceptible to damage from flying debris.

[165] I am satisfied that the explosion venting on the surface fan was inadequate and failed to protect the fan from the effects of the overpressure from the explosion.

[166] Even without specific guidance, the company should have seen on any review that the venting would have revealed it was well undersized. Also that the fan was located too close to the shaft and the control panel too close to the access door. The location of the surface fan and its inadequacy were dramatically demonstrated by the fact that it was unable to fulfil its design function as backup emergency ventilation.

[167] During the course of evidence, Mr Reece informed me that until quite recently in Australia, there was legislation that stipulated that a ventilation fan for an underground mine could not be located underground. This was because of the inherent dangers and difficulties of locating a ventilation fan underground. This is no longer a requirement of the legislation as no-one in Australia would now locate a ventilation fan underground.

[168] At Pike River Coal, the main ventilation fan was located underground. Prior to the development of the mine, there were various feasibility reports as to the location of a fan. Some of them proposed a surface fan positioned at a short fan drift at the top of the ventilation draft. However there were difficulties with that given access and the geology of the position. Access would only have been afforded by helicopters and this would have been dependant on weather conditions. This would also have compromised ongoing maintenance of the fan.

[169] Part of the decision to site the main fan underground was the potential to have a backup fan on the surface. The idea was that if the underground fan failed, then the surface fan could take-over the ventilation.

[170] There was, however though, a problem in that all of the electrical services for both fans ran through the mine. This meant that in the event of methane in excess of

1.25 % being present in the region of the fan motor, electricity would be tripped to the portal substation and the main fan would cease to operate. It also meant that electricity could not be used by the surface fan to degas the mine at any time when methane in excess of 1.25 % was present underground and any degassing would

have to rely on the backup diesel generator. The backup fans were generally only accessible by helicopter, again this being dependent on weather. Foot access was possible but extremely challenging.

[171] In 2007, another option was discussed of using a forcing fan sited at the main portal. Although this is generally used in hard-rock mining, and is unusual in a coal mine, a fan located at the portal would have been accessible at any time and in particular, after an explosion. This also could have been installed with conventional explosion protection to maximise its chances of ongoing operation after an explosion. Fans such as these are currently used to ventilate coal mines in Canada.

[172] Despite the geographical difficulties to the site, I am satisfied that the placement of the main ventilation fan underground was not the safest option available to the company due to the real possibility of damage from the explosion, the constraints on power supply in a gassy atmosphere and the inability to guarantee access to the backup service fan at any time. The company had been informed of the alternative options; however, they still went ahead with the underground installation. Although I find that the company breached this practicable step, I do not see this as having a causative effect in relation to the explosion that occurred on 19 November. The difficulty has been, however, that re-entry to the mine has been significantly compromised.

[173] After making the decision to site the main fan underground, the company failed to re-evaluate the level of risk and ongoing implementation of controls associated with this decision. The company encountered a number of problems positioning the fan in its original placement. The original design and risk assessment was undertaken in February 2007. A contract for the supply of the fans was finalised in 2008. This relocated ventilation shaft which was originally suggested approximately 2.3 kilometres from the portal and to the west of the Hawera fault. This shifted the fan chamber from solid rock into a coal measure. This introduced additional risks from the liberation of methane and spontaneous combustion. No new

risk assessment or review of the earlier risk assessment on siting the main fan underground was undertaken.

[174] The initial ventilation shaft began on 15 December 2008. However, the lower part collapsed in January 2009 and another shaft (called an alimak raise) was driven adjacent to the collapsed section of the shaft. This meant that the installation and return headings had to be redesigned. No new risk assessment or review of the earlier risk assessment was undertaken.

[175] Advice was further received in April 2009. This suggests that long-term a 1.6 stone drive would serve as an explosion path. This would minimise explosion damage by focusing on limiting coal dust explosions through the use of a stone dusting and explosion suppression barriers. In June 2009, the company deferred the excavation of the 1.6 stone drive. It noted that other methods would be used. These were for example, stone dust barriers. Although stone dust barriers were purchased, they were still present in the storeroom on 19 November and had not been installed underground. There had been no review of any further risk assessment of the placement of the fan or the decision not to develop the 1.6 stone drive.

[176] On 31 March 2010, the company received a report from the commissioning engineer for the main fan. He recommended that another risk assessment be undertaken in relation to the proposed design and installation of the fan. No review or further risk assessment was undertaken. The board notes that on 24 August 2010 the deferral of the stone drive was again discussed. This was again discussed on

13 September and 24 September. In that last meeting it was confirmed that the second return road which could be eliminated with a net saving to the company of

5.5 million. None of the meeting minutes or the presentation on 24 August discussed the use of the stone drive as an explosion path. There was no review or further risk assessment of the placement of the main fan underground.

[177] At the end of October 2010 the company began taking samples from underground headings to confirm that sufficient quantities of stone dust were in place to render coal dust inert in an explosion. Up until that time, no samples had been taken. The results from these samples confirmed that none of the headings

tested met the requirements of Regulation 36 of the Health and Safety in Employment (Mining Underground) Regulations 1999 and that all of the samples contained too much combustible materials.

[178] At the time of the explosion on 19 November, the underground fan was in the direct and only path of the explosion overpressure and there were no explosion suppression barriers in place underground to limit the possible coal dust propagation of the explosion flame front. As a result of the explosion, the fan ceased to operate.

[179] Due to the difficulties already discussed in relation to the surface fan, no ventilation has been restored to the mine since the explosion on 19 November. I am satisfied that the company failed to take a practicable step in terms of continuing to assess the risk for siting the main fan underground and that the risks should have been revisited at every significant change to the siting of the fan and the changes to the proposed controls that had been suggested and were to be implemented.

[180] As it was, by 19 November, almost all of the original controls proposed in the original risk assessment had been deleted and the main fan was unprotected from an underground explosion.

Practicable step 4 – Minimisation of the risk and damage to the underground fan

[181] This practicable step, in my view, is covered by practicable step 3 under this heading and therefore, I find that I do not to consider it, having found the company has breached its obligation under practicable step 3.

[182] Water or stone dust barriers are designed to extinguish a flame front of an explosion. Therefore, they prevent the propagation of an explosion through an underground mine via the ignition of coal dust. The installation of barriers is explicitly required pursuant to Regulation 36(c) of the Health and Safety in Employment (Mining Underground) Regulations 1999. This was also reflected in the company’s ventilation management plan which stated:

Stone dust barriers will be used at Pike Coal Ltd’s mine. Barriers shall be installed to cover all possible paths of an underground ignition and each ventilation split where coal cutting operations are conducted shall have independent explosion barriers.

19 November they were in the storeroom. There were no stone dust or water barriers installed underground. There were no independent explosion barriers at any ventilation split and in particular, where the coal cutting was being conducted. The company had already gone to the expense of purchasing the barriers and it would have involved little cost to install them. In the event of an explosion, these barriers, (Mr Reece confirmed during the course of his evidence) are very effective in stopping any flame front from propagating through the mine. The company should have installed the barriers to reduce the effects and extent of an explosion.

[184] In a mine it is literally “pitch black”. Both Mr Reece and Ms Birdsall confirmed that in a mine with no lights, you cannot see your hand in front of your face. Therefore, smoke lines are essential for personnel to follow when there is no lighting available. A smoke line is clipped to the roof of the mine and a bungee cord is attached to it and when one pulls the bungee cord it releases the smoke line from the roof to approximately head height. The company proposed a standard for positioning cones and fluro droppers which are part of a smoke line system in its emergency evacuation of underground mine report dated 24 March 2010. Despite this, there was no evidence in the documents supplied by the company of a standard operational procedure for setting out the installation and the maintenance of smoke lines.

[185] On three occasions in 2010, the condition of a smoke line was recorded in minutes of the PRC Health and Safety Committee minutes. Internal mine audits commented unfavourably on the condition of smoke lines. Interviews conducted post the explosion noted that the smoke lines were either absent, incomplete, on the floor, or at a height where they were difficult to reach. As late as 17 November, a deputy report noted “smoke line in very poor condition, needs to be looked at”.

[186] Due to the disorientation in a mine and the loss of visibility after an explosion, smoke lines are likely to be extremely important to provide an essential safety function by leading personnel to safe refuges.

[187] I am satisfied that the company failed to take all practicable steps to install functional smoke lines in all the headings of the mine. It, therefore, placed its employees and persons legitimately in the mine at risk.

[188] A FAB is a fresh air base. It is used as a place of refuge by personnel underground until they can safely retreat in the event of an emergency. All miners have access to a self-rescuer, which will only last for a limited amount of time. An FAB should provide long-term refuge for personnel who cannot evacuate the mine.

[189] The self-rescuers supplied at the Pike River Mine generated sufficient oxygen to survive in an irrespirable atmosphere for at least 25 minutes and were worn on the belt by everyone while they were underground. FAB have certain requirements:

Protection from irrespirable atmosphere including a provision for entry through an “airlock”.

Isolation from combustible materials in floor, walls and ceilings. Isolation from additional fuel sources such as methane drainage. A robust means of communication with the surface.

[190] There was only one FAB at the Pike River Mine. This was located a short distance from the ventilation shaft which required a person to climb a series of vertical ladders over a 100 metres. The other alternative was to evacuate through the main 2.3 kilometre portal tunnel. The FAB was inadequate. It was excavated in coal and was not lined with incombustible products. It was only sealed from the mine by

a roll-down brattice cloth curtain. There was no explosion protection and it did not provide an airlock for safe entry. Entry through the curtain could lead to contamination of the atmosphere in the FAB. Fresh air was supplied through a slim- line shaft. This flow relied on the lower air pressure underground. This could easily be disabled due to an explosion. There was no backup fan or pump in the FAB to continue to supply fresh air if the main fan stopped (as it did). The seal on the brattice curtain also relied on the deferential pressure created by the main fan’s operation. Once that stopped, or following an explosion, it’s possible that would have been reversed and the brattice curtain would not guarantee that this area would remain sealed.

[191] Importantly, the methane drainage line ran through the roof of the FAB. This contained almost pure methane. If this line was damaged at all near the FAB during an emergency, methane would be leaking directly into the FAB and would be an explosive fuel source as it would be mixing with fresh air. The FAB was inadequate in its size and it would not comfortably accommodate all of the personnel who may have been underground. There was only one line of communication to the outside world, and that was by a phone line that ran down the main drift, with all the other underground communications.

[192] I am firmly of the view that the FAB 2 at Pike was inadequate for any of the purposes it was specifically designed for. I find that the company failed in relation to this step.

Practicable step 8 – PRCL should have implemented an effective system to know the whereabouts of all persons

[193] It is basic commonsense that in the event of an emergency, the company should know how many people were in the mine and where they were located. Not only in relation to its own employees but also in relation to contractors. This makes it easier for rescue personnel to locate potentially injured miners or contractors, and also allows for rescuers themselves to not be put in danger searching areas of the mine where there are no personnel.

[194] Regulation 15 of the Health and Safety in Employment (Mining Underground) Regulations 1999, requires that an employer or a person who controls the place of work, must take all practicable steps to ensure that an adequate record is made of entry and exit from the mine of every employee, including contractors and employees of contractors, and that a record is kept at the entry point. On

19 November there was a tag board which recorded the entry and exit of people from the mine. This was located on the ramp outside of the lamp-room, rather than at the portal or entry point of the mine. This tag board was misleading given its location, as on

19 November there were at least two tags on the board of persons who were not underground, and at least one tag was missing from the board from a person who was underground. As a result, it took a considerable period of time to clarify exactly how many men were underground and who.

[195] The company was aware of the difficulties and the inconsistency of the use of the tag board. Both in June and February 2010, the company was aware that the tag board procedures had not been followed. In August and October, two further incidents were noted.

[196] Despite these problems, the company did not enforce the requirements of the tag board and did not audit compliance with the tag board system. In addition, once in the mine, the company had no means of ascertaining the movements of persons, either employees or contractors, within the mine. There was no method either within the mine or on the surface to systematically track and record the location of people underground.

[197] There was no means of tracking the whereabouts of the various contractors within the mine and there was no record available, for example, to locate three fitters, being Malcolm Campbell, Peter Rodger and Jacobus Jonker. There were general instructions that the deputy should be ringing in to the control room every two hours. However, this was not always followed. There are no recordings from

19 November or any report about who was in a particular district underground. On

19 November, there was no manning sheet provided to the control room operator so that the company was aware which employees were manning which vehicles.

[198] There was no evidence of a system which required contractors to directly inform the control room operator of their expected locations and the work that they were doing before going underground. They were expected to sign in and out of the visitor and contractor sheet held at the control room as they were entering underground. However, this was not always complied with. Although the contractor had access to telephones underground and a communication system called DAC, a review of the transcript from 19 November shows that contractors were largely silent, did not report their movements through the mine. There were no rules that restricted contractors to only certain parts of the mine. It is common practice to restrict contractors and workers to certain districts in a mine. Such practices as allocating colour-coded hats can be used as a help to enforce this.

[199] I am satisfied that PRCL should have implemented a rigorous system to ensure that it had knowledge of the whereabouts of all personnel underground that was up-to-date and accurate. The company’s breach of this fundamental obligation has exacerbated the difficulties and dangers associated with the recovery operation. This was a fundamental step to enhance the safety of those underground after an explosion, and so crucial to the safety of any rescuers.

[200] The informant relies on a failing to inspect its ventilation control devices and that there were inadequate systems in place to record and audit these ventilation control devices. I have already found that the company breached its obligations in this regard in a fundamental way. In my view, in relation to Mr Dunbar, the company also failed in a number of other respects.

[201] Of particular relevance in relation to Mr Dunbar: I find that the inadequate monitoring of the gas recording devices, the inadequacies of ensuring or mitigating the risk of ignition sources, inbye the panel as being significant breaches that were causative of the explosion and hence Mr Dunbar’s death. Therefore, Mr Dunbar and his family members are victims as defined in s 4 of the Victims Rights Act 2002.

Term	Common definition
Absorption Isotherm	A theoretical measure of the potential gas that could be contained within a coal inferred to be in its natural state.
Air short circuiting	When all or part of the airflow does not complete the planned ventilation circuit because it finds an alternative route.
Anemometer	Instrument for measuring the air velocity within roadways.
Auxiliary fan	Smaller fan used to ventilate dead-end roadways underground. Used in conjunction with ducting to force or extract air to the end of the road.
Bleeder heading	A roadway that is a part of the ventilation system used to manage methane accumulations in a goaf. Return roadway on the downwind side of an extraction area that is not directly connected by a roadway to the intake (fresh air) side. Used to draw seam gas away from the extraction area – carries the contaminated air.
Booster fan	Fan located underground within the main ventilation circuit to increase airflow. The fan is installed so that all air passes through the fan.
Bore hole/Drill hole	Drill hole created by drilling to gather geology information or gas drainage. Can be done from the surface or underground.
Brattice Fire Resistant Anti Static (FRAS)	Impervious plastic/fabric cloth used in the construction of ventilation control devices, eg stoppings, curtains etc.
Caving/roof fall	Process where the roof is undermined and fails to the extent that the roof collapses.
Coal measures	Coal measures comprise strata containing coal seams deposited in the same geological period.
CO/CO₂ration	Ration of carbon monoxide to carbon dioxide concentration used to assist the assessment of spontaneous combustion. A high proportion of CO is indicative that spontaneous combustion is occurring. Typical ratio are: < 0.02: normal; < 0.05: temperature of coal < 60oC; < 0.10: temperature of coal < 80oC;

	< 0.15: temperature of coal < 100oC.
Control room	Surface location (operations centre) performing the centralised function of monitoring, operating and controlling the mine. This involves items such as data recording, controlling pump and conveyor systems, monitoring the mine atmosphere and responding to alarms. The control room acts as central communication point as is typically manned when personnel are underground.
Crosscuts or cut-through	Underground roadways developed at regular intervals to join one or more main roads.
Desorbed gas	Desorbed gas is that gas directly measured from a coal sample, normally based on a coal core cut during surface or inseam drilling.
Descensional ventilation	Movement of air to a lower point (downhill).
Dedicated bore holes	Drill hole with a single use/purpose.
Designated air	Marked location underground.
Drift/drive/tunnel/adit	Roadway driven in stone. Alternative definition(s): A tunnel* or drive is an underground passageway, completely enclosed except for openings for egress, commonly at each end.* A drift* is an inclined access from the surface to the coal seam or from coal seam to another coal seam.*
Explosion doors	Protection device in the form of hinged doors/covers on the ducting leading to the main fan that are forced open by the pressure generated by an explosion. This is done to provide some protection to the fan from the force of the blast.
FAB – Fresh Air Base/emergency refuge	Location underground where a known fresh air source is available, this may have an air source that is independent of the main ventilation air. An emergency refuge will generally have a supply of self contained breathing apparatus or refill station if using a CABA system. This was a roadway stub developed off the main access road within Pike River.
Fan and its components	CENTRIFUGAL or AXIAL FAN Housing – contains the internals Fan Internals – Impellor & blades Diffuser – duct on outlet side Ducting – connecting to mine working Motor – located next to the housing

	Control system & electrics – controls motor A mechanical device used to create the air current within the mine, drawing in fresh air and removing “contaminated or bad” air.
Fault	A fault is a planar discontinuity in a rock mass across which there has been significant displacement resulting from the action of tectonic forces. Energy release associated with rapid movement on faults is the cause of most earthquakes.
Gas content (of a coal)	All coals inherently contain gas which in New Zealand is typically methane with trace amounts of ethane and possibly propane.
Gas drainage pipes/system	Network of pipe work to reticulate the seam gas being emitted from the in-seam boreholes to a discharge point. The system may include valves, water traps to collect condensation in the pipeline and a flame arrestor at the discharge point to prevent ignition. A pump on the surface may be used to assist the gas to flow and reduce the required size of the pipe work. Alternative definition The system of boreholes, pipes and other devices to capture gas from the coal seam and surrounding strata and reticulate it to a discharge point(s).
Goaf	That part of a mine from which the mineral (eg coal) has been partially or wholly removed. Generally these areas are collapsed or partially collapsed and no longer suitable for access by people or equipment. Is also called the gob.
Graham’s Ratio	Ratio of the carbon monoxide produced to oxygen consumed and is directly proportional to the temperature of the coal or the extent of accelerated oxidation. Typical European rations are: < 0.4: normal; > 0.4: possible early stages of spon com; > 1.0: spon com event almost certain; > 2.0: serious spon com event; > 3.0: active fire.
Headings	Two or more roadways generally driven parallel to access different parts of the mine.
Hydraulic mining/hydro monitor	Process of excavating stone or coal with the use of high pressure water and specialised equipment. This

	was the planned extraction process within Pike River. The hydromonitor is the high pressure water equipment.
In seam drainage	Removal of coal seam gas with the use of inseam drill holes and possibly associated pipe work. Is a method of reducing the insitu gas content of the seam to within acceptable limits by drilling holes into the seam or surrounding strata ahead of mining.
In seam drilling	Drilling of boreholes in and around the coal seam from an underground location.
Inbye	The direction towards the coal face from any point of reference.
Intake	U/G roadways that have uncontaminated/fresh air moving through them that has not yet passed a working place.
Maihak tube bundlers or Tube bundle system	Gas monitoring system that continually draws (vacuum based) air/gas samples through a network of plastic tubes placed around the mine to the surface for analysis. There is a delay from when the sample is taken to when it reaches the analyser. SICK Maihak GmbH is the leading manufacturer of such systems.
Mains/Selection/panels	Mains: Group of roadways that provide long term people and equipment access and ventilation pathway to get to and from the mining areas (panels/sections). Section: Also known as panel. Mining area connected to the mains roadways consisting of access roads and extraction areas with a separate ventilation circuit. Panel: Also known as section. Mining area connected to the mains roadways consisting of access roads and extractions areas with a separate ventilation circuit.
Methane outburst	The sudden ejection from the solid coal face into the mine workings of methane, carbon dioxide and generally including coal and rock.
Mine gases	Carbon Monoxide (CO): A colourless, odourless gas, CO, formed by the incomplete combustion of carbon or a carbonaceous material (eg diesel machines, mine fire, spontaneous combustion of coal). Highly flammable (12.5 to 74%) & very toxic in low concentrations. 25 PPM No effect (TWA) 1000 PPM – headache, nausea, and dizziness. Vomit after 30 mins 1500 PPM – possible collapse after 15 mins

	3000 PPM – immediate physiological effects, unconsciousness after 5 mins. Carbon Dioxide: 0.03% of Air. Formed U/G by engine exhaust, oxidation of coal or fire. May be a coal seam gas. Colourless but pungent odour. 0.5% - slight increase in respiration (TWA) 2% - 50% - increase in respiration 3% - (STEL) 5% - 300% - increase in respiration 10% - intolerable. Hydrogen (H₂): Colourless tasteless and odourless. Highly flammable (4 to 74 %). May be produced as product of spontaneous combustion. Hydrogen Sulphate (H₂S): colourless gas with rotten egg odour. Flammable (4.3% to 45%). 0.1PPM: detected by smell 10PPM: TWA 15PPM: STEL < 100PPM: irritation to eyes and respiratory tract. Methane (CH₄): coal seam gas. Colourless tasteless and odourless. Highly flammable (5.0 to 15%). auto ignition (that is, will burn) temperature 537oC. Ethylene (C₂H₄): Spontaneous combustion indicator – detectable only with Gas Chromatograph. Presence in low quantities indicates temperature of 100 – 150oC.
PPM	Parts per Million
STEL	Short-term exposure limit means the maximum average exposure (to the gas) measured over any 15- minute period in the working day. Measurements over the STELs will require removal of personnel from the exposure area.
Monitoring non-restricted environment/non restricted environment qualification	Defined by Mining Regulations (Underground) 1999. Restricted zone means – (a) all parts of a ventilation district in a gassy mine that are on the intake side and within 100 metres of – (i) the most inbye completed line of crosscuts; or (ii) a longwall or shortwall face; or

	(b) a part of a gassy mine in which flammable gas, whether or not normally present, is likely to occur in such quantity as to be 2% by volume or more in the general body of air in the gassy mine; or (c) a part of a gassy mine in which electrical equipment is located and that has not been shown to be free from flammable gas; or (d) all of the return side of a gassy mine. The Regulations stipulate additional equipment restrictions, record keeping and air management standards for restricted zones.
Outbye	The direction away from the coal face from any point of reference.
Overcast	A structure build in an underground roadway intersection to keep air paths separated. An air crossing between intake and return roadways.
Oxidation	The reaction of coal with oxygen that produces heat and gas. The rate of oxidation is affected by the surface area of coal, coal type, temperature and available oxygen.
Pit bottom	First area to be developed in the coal seam where underground services are established eg the area of initial mine development at the end of the stone access roadway within Pike River.
Portal	Surface entry of a main roadway into the mine.
Raise bore	Method of developing a shaft where initially a pilot hole is drilled from the surface to an existing roadway. A drill head is attached underground and is pulled/rotated back to the surface to enlarge the hole.
Recirculation (ventilation)	When return, entrained or exhaust airflow including contaminants re-enters the incoming ventilation path.
Return	Any underground roadways that have “used contaminated” air moving through them towards the surface after it has passed a mining area.
Self-contained self- rescuer/self-rescuer/SCSR	A temporary breathing system for use when the mine atmosphere becomes unbreathable. There are two possible systems: one with a simple filter (rarely used); the other, using potassium peroxide, reacts with exhaled CO₂and produces sufficient oxygen for approximately 30 to 60 minutes of use. Intended to allow the user to move from current location to fresh air or other air source. Alternative definition:

	Escape breathing apparatus shall be classified in accordance with the following types: Chemical Oxygen Self-Contained Self Rescuers (chemical oxygen apparatus) Carbon monoxide filter self-rescuers Compressed Air Self-Contained Breathing Apparatus (compressed air apparatus) Compressed Oxygen Self-Contained Breathing Apparatus (compressed oxygen apparatus)
Shaft	Vertical access way between two points connecting the surface with the underground workings eg 100 metres long and 5 metres in diameter.
Smoke lines	A series of rope lines and small cones hung along underground roadways to assist in guiding people through the mine to a point of safety in the event of an emergency and low visibility.
Spontaneous combustion (also termed a heating when coal is involved)	Coals reacts with atmospheric oxygen even at ambient temperatures and this reaction creates heat. If the heat liberated during the process is allowed to accumulate, the rate of the above reaction increases exponentially and there is a further rise in temperature. When this temperature reaches the ignition temperature of coal, the coal starts to burn and the phenomena is described as spontaneous combustion.
Tunnel	Horizontal roadway primarily driven in rock that links the surface operations to the coal seam. An underground roadway could be used along with similar terms as: roadway, drift, heading.
Up-dip	Pump that generates high pressure and high volume water that is used to excavate coal via the hydromonitor.
Upsidence	The upward rupture of the beds of incised streams as a result of a wider area of subsidence such as results from coal mining.
Ventilation pressure	Pressure required to overcome the resistance of air moving through the mine.
Ventilation circuit	Pathway that air follows through the mine or section/panel of the mine.
Ventilation system	Roadways and equipment required to direct, control, push, pull air throughout the mine. This involves fans, ducting, artificial walls etc. *Alternative definition* Describes the arrangement of facilities (including fans

and other ventilation control devices), and the distribution of air throughout the mine roadways and workings that provide for a sufficient volume of airflow to remove heat and to dilute and render harmless noxious, and flammable gases, and dust.

District Court of New Zealand

Department of Labour v Pike River Coal Limited DC Greymouth CRN 11018500202 [2013] NZDC 807 (3 May 2013)