Morrison v Powercoal Pty Ltd

PARTIES : DEFENDANT IN IRC6372 AND 6376 OF 2001 Powercoal Pty Limited

DEFENDANT IN IRC6375 AND 6377 OF 2001 Peter Lamont Foster FILE NUMBER: IRC6372 and 6376 of 2001 CORAM: Peterson J CATCHWORDS : Occupational health and safety - s15(1) and s50(1)- Collapse of roof in coalmine - Pillar stripping in second workings - Indicia of poor roof - Whether offence made out as to system for assessing the safety of the roof - Whether offence made out in relation to a failure to record and notify faulty roof discovered in first workings - No offences made out - Charges dismissed. LEGISLATION CITED : Occupational Health and Safety Act 1983 ss 15(1) Evidence Act 1995 WorkCover Authority of New South Wales v Howard & Sons Fireworks Spectaculars Pty Limited [2000] NSWIRComm 211, 31 October 2000, unreported WorkCover Authority of New South Wales v Arbor Products International (Australia) Pty Limited (2001) 105 IR 81 CASES CITED : WorkCover Authority of NSW (Inspector Twynam-Perkins) v Maine Lighting Pty Limited (1995) 100 IR 248 Dunlop Rubber Australia Ltd v Buckley (1952) 87 CLR 313 at 320 Smithwick v National Coal Board (1950) 2 KB 335 HEARING DATES: 03/17/2003; 03/18/2003; 03/19/2003; 03/20/2003; 03/21/2003; 03/24/2003; 03/25/2003; 03/26/2003; 03/27/2003; 03/28/2003; 03/31/2003; 04/01/2003; 04/02/2003; 04/03/2003; 04/04/2003 DATE OF JUDGMENT: 11/21/2003

PROSECUTOR Mr S Crawshaw SC with Ms A F Backman of counsel

SOLICITOR Crown Solicitor's Office SYDNEY.

DEFENDANT IN IRC6372 AND 6376 OF 2001 Mr B D Hodgkinson SC with Mr P Nolan of counsel LEGAL REPRESENTATIVES: DEFENDANT IN IRC6375 AND 6377 OF 2001 Mr M L Shume of counsel

SOLICITOR Sparke Helmore SYDNEY.

JUDGMENT: - 40 - INDUSTRIAL RELATIONS COMMISSION OF NEW SOUTH WALES

RODNEY DALE MORRISON v POWERCOAL PTY LTD Prosecutions under s 15(1) of the Occupational Health and Safety Act 1983

RODNEY DALE MORRISON v PETER LAMONT FOSTER Prosecutions under s 50(1) of the Occupational Health and Safety Act 1983

1 These are four prosecutions taken by Rodney Dale Morrison as a person authorised to institute proceedings against Powercoal Pty Ltd ('Powercoal') and the (former) manager of Powercoal's Awaba Colliery, Peter Lamont Foster. The prosecutions, brought under the Occupational Health and Safety Act 1983 ('the OHS Act') have been commenced following a roof collapse in the Awaba Colliery on 17 July 1998, the direct result of which was the death of a miner, Barry John Edwards. However, it is not Mr Edwards' death to which the charges go.

2 The two charges against Powercoal, in substance, are a. Matter No IRC6372 of 2001

Between about December 1996 and 17 July 1998 at the Awaba Colliery, Wilton Road, Awaba in the State of New South Wales, the defendant, being an employer, failed to ensure the health, safety and welfare at work of all its employees, in particular, Barry John Edwards, in that it failed to provide a safe system of work contrary to section 15(1) of the Occupational Health and Safety Act 1983. The particulars of this charge are extensive but include the following, which are relevant to set out:

(xi) there was a potential risk to the health, safety and welfare of the employees working at 304 Panel 25 CT, in the vicinity of 1, 2 and 3 Headings, between 15 and 17 July 1998 in that there was a risk of the roof falling in while the employees were working there;

(xii) the defendant's failure and/or omission was the failure and/or omission to provide an adequate system for assessing the safety of the roof in the vicinity of 304 Panel at 25 CT, in the vicinity of 1, 2 and 3 Headings;

(xiii) there was a causal nexus between the risk to the health, safety and welfare of the employees referred to in paragraph (viii) above and the defendant's failure and/or omission as dealt with in paragraph (xii). b. Matter No IRC6376 of 2001

Between about 23 September 1993 and 17 July 1998 at the Awaba Colliery, Wilton Road, Awaba in the State of New South Wales, the defendant, being an employer, failed to ensure the health, safety and welfare at work of all its employees, in particular, Barry John Edwards, in that it failed to provide a safe system of work contrary to s15(1) of the Occupational Health and Safety Act , 1983. The particulars of the charge are extensive but include the following, which are relevant to set out:

(xii) there was a potential risk to the health, safety and welfare of the employees working at 304 Panel 25 CT, in the vicinity of 1, 2 and 3 Headings, between 15 and 17 July 1998 in that there was a risk of the roof falling in while the employees were working there;

(xiii) the defendant's failure and/or omission was the failure and/or omission to implement an adequate system of recording and notification to employees of roof problems and roof history.

(xiv) there was a causal nexus between the risk to the health, safety and welfare of the employees referred to in paragraph (ix) above and the defendant's failure and/or omission as dealt with in paragraph (xiii). The charges against Mr Foster reflect the two charges against Powercoal and allege that Mr Foster: "being a person concerned in the management of the above corporation, is deemed to have contravened the same section pursuant to s 50(1) of the Occupational Health and Safety Act, 1983".

3 The charges against Mr Foster arise as a result of s 50 of the OHS Act. That provision is to the effect that where a corporation contravenes the OHS Act, whether by act or omission, each director of the corporation and each person concerned in the management of the corporation shall be deemed to have contravened the same provision unless the court is able to be satisfied that the person was not in a position to influence the conduct of the corporation in relation to the contravention or, being in such a position (that is to influence) used all due diligence to prevent the contravention by the corporation.

4 Certain facts in the matter were agreed including the prosecutor's authority; the fact that Powercoal was the operator and owner of the Awaba Colliery, at least from 23 September 1993 until 17 July 1998 and was the employer of employees. In particular, on 15, 16 and 17 July 1998 it was the employer of Messrs Gardner (a mine deputy) and miners Ian Miles, Warren Etheridge, Kevin Dunn, Alyn Gear, Mr Edwards and also Greg Cornwell, an electrician who also undertook relief mining work.

5 Other background facts which I find are that the defendant Peter Foster was, on 23 March 1998, appointed the mine manager of the Awaba Colliery pursuant to s 46 of the Coal Mines Regulations Act (CMRA).

6 In the course of the prosecution's case evidence was being led in chief from Dennis Charles Dellaca, an employee of Powercoal at the time of the alleged offence. Mr Dellaca was then employed as a mining deputy, usually on afternoon shift, working in the panel in which a roof fall occurred, on day shift on 16 July 1998, causing the death of a continuous miner operator, Barry John Edwards. Mr Dellaca's evidence, at the relevant point, was dealing with his experience in 304 panel, where the fall occurred, on the previous day. In particular, he was being asked about his memory of a roof fall which he had observed on 15 July 1998. After having been given leave to refresh his memory of events by reference to a record of interview undertaken with an inspector authorised under the CMRA to investigate the accident and to compulsorily interview relevant persons to that end, leave was given to cross-examine him on the basis that his record of interview constituted a prior inconsistent statement. At the conclusion of that cross-examination senior counsel for the prosecutor, Mr Crawshaw SC, with him Ms Backman of counsel, sought to tender one sentence from the interview and part of another sentence. The sentences sought to be tendered thus are those set out in italics in the following four sentences: We had taken a couple of lifts off and experience a small fall in the area off sequence 4. I presumed it was associated with the fault marked on the plan.

After the fall, which was about two lifts wide and about 400 mm thick, perhaps thicker in parts, I checked the roof through to 3 Hdg 25 intersection. There was nothing to indicate to one that anything was wrong and that the fall was mainly a localised bit of poor roof.

7 The tender was made, against both defendants, on the basis of s 87(1)(b) of the Evidence Act 1995 as an admission. Admission is defined in the dictionary to that Act as: A previous representation that is: (a) made by a person who is or becomes a party to a proceeding (including a defendant in a criminal proceeding); and

(b) adverse to the person's interest in the outcome of a proceeding.

8 Section 87(1)(b) provides: For the purpose of determining whether a previous representation made by a person is also taken to be an admission by a party, the court is to admit the representation if it is reasonably open to find that:

(b) When the representation was made, the person was an employee the party, or had authority otherwise to act for the party, and the representation related to a matter within the scope of the person's employment or authority.

9 Objection was taken to the tender by Mr Hodgkinson SC, appearing for the defendant, Powercoal, with Mr Nolan of counsel. The first argument was that the statement was not a representation on the basis that the statement made was "an observation made by him, not a representation as to a matter within the scope of his employment". In the event that the court was against that argument it was submitted that "representation", properly understood in s 87(1)(b) does not go to a statement compulsorily required by an investigator. It was submitted that the only purpose of the tender was either a prohibited purpose - a confirmation through the statement of the evidence that has been extracted or to impugn the credit of the prosecution's witness. Senior counsel also adverted to the need to look to broader reasons such as prejudice as to why it ought not be admitted.

10 Mr Shume of counsel, appearing for Mr Foster, took the position that the representation could never be admissible against Mr Foster given the express terms of s 87(1)(b). Were Mr Foster being tried separately, the representation could not be admitted against him and the fact that a joint trial is being undertaken, does not alter that position. So far as Mr Foster is concerned, the representation would be a third party admission which would require his consent under s 83 and, in the absence of which, it would be inadmissible.

11 In reply Mr Crawshaw submitted that s 87 does not provide any exclusion in relation to compulsory interviews and the proposition that representations amounting to admissions made in that context are excluded is simply wrong. He submitted that the checking or noticing of roof falls was obviously a matter within Mr Dellaca's employment as a deputy. No issue of credit arises. Mr Dellaca had no current memory of the matters but his record of interview correctly recorded his views at the time.

12 An objection by Mr Hodgkinson to the limited form of tender of only parts of the relevant passages in Mr Dellaca's statement were responded to by Mr Crawshaw by accepting that the wider context should come within the tender. Following that argument I made the following ruling, indicating that I would provide short reasons for the ruling in due course: "I propose to admit into evidence pursuant to section 87 of the Evidence Act the following material which I extract from MFI4: "We had taken a couple of lifts off and experienced a small fall in the area off sequence 4. I presumed it was associated with the fault marked on the plan. After the fall, which was about two lifts wide and about 400mm thick, perhaps thicker in parts, I checked the roof through to 3 Hdg 25 intersection. There was nothing to indicate to one that anything was wrong and that the fall was mainly a localised bit of poor roof." That material is admitted as an admission by the defendant Powercoal Pty Limited, but not against the defendant Mr Foster. . . . CRAWSHAW: May I just ask about the ruling? Does that encompass the possibility that your Honour put to me when I was putting my reply, namely the possibility of seeking to rely on that material by way of submission against Mr Foster? HIS HONOUR: Yes. I had thought it unnecessary to advert to that further in the circumstances. It is not admitted against Mr Foster, but whether the consequence is that it has a bearing or effect upon him is a matter that I will hear you on in due course."

13 Having now revisited the matter, I remain of the view that Mr Crawshaw's submissions on this point were correct. Section 87(1)(b) operates in respect of an employee of a party in relation to a representation related to a matter within the scope of that person's employment. The structure of the paragraph suggests that it is the employment and the scope of the employment which provides the relevant authority to make a representation which may be taken to be an admission, as defined, by the employer party to the proceedings.

14 Apart from the reference to the possibility of prejudice, there was no particular prejudice identified which would cause me to consider the powers provided to the court by ss 135, 136 or 137 of the Evidence Act.

15 There was an obvious feature adverse to Powercoal's interest involved in the statement made by Mr Dellaca that a roof fall had been experienced in the relevant period the subject of the charges. That evidence was perhaps no more than a brick in the wall being sought to be constructed as the prosecution's case, but it was, of course, admissible for that purpose.

16 The structure of the section also provides, on my view of it, a ready answer that, within the context of a joint trial, the evidence may be taken to be an admission against Powercoal but not against Mr Foster. As Mr Shume submitted, the section refers to the representation being taken to be an admission against the relevant party not, in effect, any party. On that basis, it seems to me beyond debate that the admission was not directly admissible against Mr Foster. Whether or not that had consequences adverse to his interest should the admission have a significance in a finding of guilt by the corporation, given the operation of s 50 of the OHS Act, is a matter I left open for debate but now requires no consideration.

17 On 15, 16 and 17 July a team, consisting of the persons mentioned in par. [3], was working on day shift in 304 panel between 1 and 3 headings inbye (that is, towards the mining face from) 25 cut-through for the purpose of extracting two "pillars" of coal which had remained from the first workings, when the cut-throughs and headings forming 304 panel had been mined some years before.

18 Headings are the primary drives undertaken into the coal deposit by the continuous mining machines. Cut-throughs are the cross-drives later performed at an angle, most commonly 90 degrees, to the headings, thereby forming a squared pattern of roadways which retain their numerical nomination as headings and cut-throughs. The intersection of these roadways forms pillars of coal surrounded by the roadways. The formation of headings and cut-throughs, leaving residual pillars, is called in the industry "first workings". In the present matter the mining which was relevant to these proceedings was not the mining of a pillar isolated on all sides but two areas of coal isolated, to an extent, on only three sides. They were, in effect, on the edge of the area mined in first workings.

19 Pillar extraction occurs when the pillars are mined, and are commonly referred to in the industry as second workings. There are two methods employed in the industry. Pillar extraction is the term usually applied to the mining of the whole of the pillars. In this case the programme for mining 304 panel was not complete pillar extraction but what is referred to as "pillar stripping". This involves the extraction of approximately half of each pillar.

20 The objective of pillar mining or extraction is principally to obtain the residual coal in the pillars, but in a manner which involves retreating from the mined area, in effect towards what I might loosely call the exit, thereby intending to ensure that, while working, the miners remain under a roof structure which has been appropriately supported, usually by roof bolts, during first workings. It appears to be a fundamental tenet of mining practice that miners and supervisory staff must not perform work under an unsupported roof because of the inherent risk involved of a roof collapse. An integral part of this mining process is to leave a pillar, reduced in size as called a "stook" as a minimum support for the roof alongside a heading.

21 The mining process was being undertaken utilising a large continuous mining machine of some 15 metres length with a 3.6 metre wide cutting head or drum. The continuous miner was operated remotely by the operator standing towards the rear of the machine on its right hand side. The machine conveys the coal cut by the head to a rear-mounted conveyor which is capable of swivelling from side to side to an angle of approximately 90 degrees and which delivers the coal to a shuttle car driven by another miner. A shuttle car upon receiving a load then transfers the coal to a major conveyor system some distance away. Two shuttle cars were used for this purpose, one loading while the other transported.

22 The coal being mined formed part of what is called the Great Northern Seam, in this case near the western shore of Lake Macquarie, south of Newcastle. The coal seam at 304 panel was approximately three metres thick or high and was lying between what is called Awaba Tuff, a substance now accepted to be the compacted deposits of volcanic ash and, above it, what is called Teralba Conglomerate.

23 The geological nature of "conglomerate" is dealt with in the report in evidence of Dr Moelle, to which I shall later refer. For present purposes it is sufficient to describe conglomerate as a layer of hardened substance comprising stone pebbles and some binding material and is particularly hard. After the mining process the conglomerate effectively forms the roof of the mine in that area. Accordingly, after the headings and cut-throughs have been mined in an approximately squared formation of tunnels, the layer of conglomerate lies across the mined area and the remaining pillars, forming a more or less continuous roof, supporting the terrain above and supported itself by the pillars.

24 The thickness of the conglomerate has an effect upon its strength and ability to bridge open areas. There are, however, faults in the conglomerate where there has been a shearing of the terrain in a vertical fashion such that the conglomerate does not form a continuous layer but can be either lowered or raised by the shearing effect; there was such a fault of 0.3 metres in 304 Panel and another, nearby but in an area not to be mined up to eight metres.

25 One issue in the proceedings relates to the nature of the roof itself, the prosecution submitting, in part, that the first workings had detected "non-conglomerate roof" and the defence submitting that, on the evidence, it should be concluded that the roof was a conglomerate roof.

26 Headings and cut-throughs are driven to an approximate maximum width of 5.5 metres. Where pillars are extracted the effect is to create a void, referred to in the evidence as the "goaf", with a heightened risk of roof fall due to the reduced amount of support.

27 Where, as in 304 Panel, two pillars are to be mined the process is laid out on a plan which allows for the retention of the two ends of the particular pillar to maintain continued support for the roof, and to take out a measured number of passes by the continuous miner, each pass or cut being referred to as a "lift". Panel 304 was to be mined in a series of sequences, the first two of which (taken out of the originally intended order, with approval) were sequences four and five. Sequences four and five involved cutting into a solid wall of coal situated between two dead-end headings, called stubs, at 1 heading and 3 heading. There was also a shorter stub at 2 heading. The plan allowed for approximately eighteen lifts to be taken. This work had been carried out substantially on 15 and 16 July and, on day shift on 17 July, the team was to complete the last two lifts before the right hand stook, known as stook X. Stook X, had it remained in accordance with the layout on the plan, would have been the left hand edge, to a minimum of 1.5 metres thickness, of the stub end of 3 heading; that is to say, on the right hand side of the layout of 4 and 5 sequences.

28 It was during the mining out of stook X itself such that it was only approximately 20% of its originally intended size (called by one witness "robbing the stook"), that the void created by the removal of the two pillars forming sequences 4 and 5 was the subject of a major roof fall. This extended over to the point where the miner operating the machine, Mr Edwards, was struck by falling rock and suffered fatal injuries.

29 It was also established in evidence that the left hand end of sequences 4 and 5 did not have a residual stook of full size. By virtue of the needs of the continuous miner and the difficulties in straightening it up, the plan allowed for a stook of 1.5 metres minimum thickness, it seems necessarily having a wider thickness at the inner end. The stook at the left hand end, which would have formed the right hand side of the stub of 1 heading, was left by the mining process not as a complete whole, but as two free-standing pillars. In the middle area of sequences 4 and 5, which sequences measured approximately 65 metres in length, two prominent fenders were left in the back wall, which was the coal face formed by the lifts, of the area opposite 2 heading. The roof fall fell from an area more or less kidney-shaped, running from the two pillars which formed the left hand stook, towards and beyond the right hand stook some 65 metres away, with the curve forming the kidney shape on the western side of the fall around each of the two pillars which had been left in the coal face opposite 2 heading. The expert evidence adduced in the proceedings suggested that the left hand stook and the pillars just mentioned had a significant effect on limiting the scope of the fall. This was also said to be the case in the area of the foreshortened stook on the right hand end of sequence 5.

30 It is also necessary to mention that as the continuous miner proceeds through its lifts along a particular pillar, the Pillar Extraction Rules adopted for Awaba and approved by the Department of Mineral Resources provide for the installation of timber props as roof support along the area which previously formed the cut-through on one side of the void or goaf area.

31 Mr Ian Charles Anderson, the Senior Inspector of Mines in the Department of Mineral Resources, prepared a report into the accident and gave oral evidence in the prosecution case. Mr Anderson concluded that the vertical stress environment at the accident site could be considered benign. He described the typical roof for the Great Northern Seam as being conglomerate which is strong, competent and capable of spanning the voids created by pillar stripping. He referred to joints, or breaks in the rock mass, as existing in conglomerate. They can be either open or closed. Closed joints tend to have little behavioural influence on conglomerate whereas open joints can and do control the caving characteristics of the rock. Open joints permit the passage of strata water and over time become discoloured with an iron stain. Discoloured joints, often referred to as "water cracks", are a sign the conglomerate is likely to be less competent than expected and to fail in non-typical fashion. The joints in the area of the accident showed no discolouration and appeared to be closed in nature. Mr Anderson believed they played no material role in creating the fall.

32 Mr Anderson's report draws the following conclusion: I conclude that over the accident site the usual conglomerate roof did not exist. A laminate bed, interlaced with coaly partings, entered the immediate roof and plunged rapidly from outbye to inbye, effectively creating a wedge of conglomerate. The interface between the laminate and conglomerate formed a significant plane of weakness in the roof. Further, at or close to the face the conglomerate wedge thinned to nothing. In the vicinity of the disappearance of the conglomerate, joints were open, further weakening the inbye end of the conglomerate wedge.

The transition from thickish conglomerate to laminate was almost certainly detected when the stub of 3 heading was driven several years ago. This explains, I believe, why the miners at that time installed such comprehensive roof support, that is roof bolts through W straps, which are absent elsewhere near the accident site.

33 The photographs taken by Mr Anderson of the area of the accident were relied on by him to conclude that the particular circumstances existing in 3 heading 25 cut-through would have, in all probability, defeated the capacity of a 1.5m bolting pattern to adequately secure the roof (this referring to the length of the bolts installed in 304 Panel). The photographs showed: ".... bolts, with approximately 0.3 - 0.5m encapsulation beyond the interface, have been either stripped of their anchorage or have had their surrounding rock and encapsulation pulled away. In effect they have failed to support the roof."

34 Mr Anderson dealt with the skimming of the stook and thought that it appeared likely the stook was being reduced in size at the time of the fall. He stated that 64% of all roof fall accidents in pillar recovery occur at intersections, the stability of which may be influenced by the stook of coal left to separate it from the extraction void. Here, the formation of the stook did not comply with the Manager's Rules for Pillar Stripping.

35 Mr Anderson dealt with the concept of a roof lithology plan in the context of his view that the laminate was almost certainly identified when the stub of 3 heading was driven. His report stated: During the design of extraction for 304 panel it would have been desirable for the manager to have a lithology plan before him outlining areas of known or suspected poor roof. In these areas extra support could be installed, stooks designed in a more conservative fashion and officials and workmen alerted to the possibility of unpredictable roof behaviour.

Formulation of such a plan must be linked to information gained upon development, on a shift by shift basis. Such information on roof type should than (sic - then) be consigned to a plan which can be used in subsequent operations.

36 In a section of his report headed "Adequacy of the General Rules of Pillar Stripping Plan" Mr Anderson stated: Plan AW 1076 is the document by which sequencing of the stripping operations in 304 panel is established. In my view there are aspects of the plan that need enhancing for the sake of clarity and safety.

There is a statement on the plan "Stook size 1.5m minimum" and arrows referring to stooks. It is not absolutely clear what the 1.5m is referring to; is it the portion of the stook next to the arrows? or is it the thinnest portion of the stook?

Further, whilst all fenders are arrowed to show direction of extraction they are not individually numbered, hence it may be possible for an operator to confuse (or possibly justify variations to) the sequencing that is permitted within the terms of the plan.

Such matters may well apply to all extraction plans within the State and hence development of guidelines for the preparation of such plans, incorporating minimum essential standards, is desirable.

37 In cross-examination relating to stook X and its effect had it been as planned, he said: ". . . what I can say is that if the stook had been as I believe it should have been, then the probability of the fall and the goaf extending into the workplace would have been very small indeed".

38 He also said in answer to a question concerned with the effect of leaving stook X as planned: ". . . certainly the extraction sequence as I read it indicates that once that last lift was made, the fender formed, they should have extracted and presumably no fall would have occurred, and if it had, the stook may well have stopped it from going to the workplace. The probability then that they would have been injured at that time would have been much, much less. So I agree with the proposition that pulling out and moving away, the risk of a fall, injuring those people at that time was very, very small".

39 He also agreed that it was possible the fall may not have occurred had stook X been left in its planned form. He also gave this evidence: Q. Of course, this retreat mining system which you described yesterday in your evidence is based upon the fact that you leave stook X then to allow time to withdraw from that working area before anything happens? A. That's one of the reasons stook X is left. I wouldn't say it's the primary reason. Stook X is left there to ensure that any goaf fall, whenever it occurs, does not run through into the intersection of the workplace where people may be during the mining process or through making that area safe. That's one of the reasons for stook X. It also provides support for the intersection during the mining of the last lift. So it's got multiple roles.

40 Mr Anderson agreed that skimming 1m or more off the left side of 3 heading, widened the roadway beyond the standard width provided in the legislation of 5.5m. This would need an exemption under the legislation, together with specific support rules approved by the Department. It was inconsistent with the support rules.

41 Mr Anderson agreed that the W-straps in 3 heading would immediately indicate to a competent mining official that there was an area of bad roof.

42 When cross-examined about the suggestion in his report that the stooks could have been designed in a more conservative fashion, he disagreed with the proposition that stook design was intended to cater for the possibility of poor roof. He contended that the design related to the normal expectation of competent conglomerate roof.

43 Mr Anderson accepted an estimate of 2,500 to 3,000 tonnes weight of the roof fall. He agreed that stook X as designed would have supported a fall of that weight, if the conglomerate sat on top of the stook. He accepted the full weight of the rock in the goaf did not sit on the top of stook X. Stook X as designed would have borne only a smaller percentage of that weight.

44 He said in relation to stook design: "They have to be based around experience and practice and a degree of conservatism because the exact loading environment is unknown."

45 In expressing the view that the review of the stook size to produce a possibly larger stook, Mr Anderson relied upon the existence of the W-strap in 3 heading and the fact that the laminate band had become apparent at the face. He agreed that he could not be certain that the laminate band was apparent before the date of the incident.

46 The court received evidence from a number of the members of the mining team that were working at the time of the roof fall. As I have said, Mr Edwards was driving the continuous miner. Mr Kevin Dunn was driving a shuttle car and was located behind the miner sitting in a forward facing seat on the left side shuttle car when the roof fell only a matter of feet in front of him. Mr Gardner, the deputy, was standing back in the intersection of 25 cut-through and 3 heading, between some timber props which had earlier been inserted in that area.

47 Mr Miles had been employed as a coalminer at Awaba for approximately 22 years. He had previously worked at Liddell Colliery. He had been employed for approximately 5 years as the local check inspector for Awaba, appointed as such by the United Mineworkers Federation, which gave him the role of particularly observing any unsafe features and reporting or repairing them. He would assist others, if asked, concerning questions of safety which they would raise with him. He held a mine deputy's ticket.

48 In cross-examination he agreed that the dribbling roof was not in the work areas, having regard to the fact that the miners were retreating back from the area in which the dribbling was occurring. He attested to there being water cracks in the roof in the vicinity of their work on 15 July. They were then working on the first pillar in sequence 4. There was a little dribbling from the roof, by which he referred to small dribbles of conglomerate in the form of pebbles coming out of the roof. He said that on 17 July, after starting at 7am, it took up to 55 minutes to get to 304 Panel. The team met in the crib room where the deputy spoke about the area and then left to make an inspection of the face. They were commencing mining at approximately half way down the second last lift on sequence 5. Mr Miles also inspected the face. He thought the job was a bit dubious so he sounded the roof around the miner. He was worried about water cracks and dribbles of pebbles and so he checked to see if there were any abnormalities in the roof. Mr Etheridge, then working as the shiftman setting timbers etc, went to remove two props on the left hand side of the miner. Mr Miles stopped him and said "I want to sound the roof first". This he did by striking the roof with a drill steel. This is a technique constantly in use by underground miners to test the roof condition. The sound caused by the steel will inform of any poor conditions in the roof. In this case, the roof sounded good, and Mr Etheridge knocked out the props.

49 Mr Miles said the first work allocation on the shift was Mr Gear as the miner driver, Mr Edwards as the cable hand, Mr Etheridge as shiftman, Mr Dunn in the right hand shuttle car and himself in the left hand shuttle car. Mr Gardner was the deputy. Also present in the team was Mr Cornwell the electrician, Mr Dymock the fitter, and two contract fitters, Messrs Heath and Kennedy.

50 The team did two shuttle cars which took about 10 to 15 minutes, then the scrubber came off the continuous miner, and required a repair for about one hour. Upon recommencing, the inbye light was out in the shuttle car and this then needed replacing. While this was happening Mr Miles went to morning tea and when he returned production was underway. During the morning, Mr Miles remained concerned about the roof and its having dribbled. He asked Mr Gear to stop the miner so they could listen to the roof and formed the view that the roof had settled down. He made very many soundings of the roof; he was quite uneasy, there was something wrong but he could not pin-point it. He said the roof was not doing what it would normally do.

51 It is instructive to record his evidence describing listening to the roof: "Well, when we have the fall, the majority of the time, because of the pressure on the strata, you can hear it cracking and working and observation of watching the roof where its dribbling and whether there's any movement there . . . . We observed for a short time that it had settled down, so we proceeded in cutting again."

52 As to the dribbles from the roof he said: Its where the water crack is and because of the roof moving, its pebbles and sometimes it comes out in - depending on how much it dribbles, it comes out most probably in handfuls or it could be just a few pebbles of stone and that falling down.

53 He later went to crib. At that time stook X was still intact. He heard the "whoof" of the roof falling while he was in the crib room.

54 In cross-examination Mr Miles stated that he observed props being put into 25 cut-through as various lifts had been completed and they didn't appear to have any weight on them. He agreed he was looking for signs of weight because he was being extra careful. He spoke of his testing of the roof and the obtaining of a clear ring when sounding it. Prior to his crib break he was present while the last lift was being undertaken and the miner was five or six metres into that lift. He said the stook was not being undersized by that process. He said the dribbles of pebbles from the roof were some six or seven lifts back from the last lift which put the dribbles approximately in the middle of the two sequences, near the goaf edge, being the area which had previously been the inbye side of 25 cut-through.

55 He agreed that when he returned to the area after the fall there was not much left of stook X, it having been mined out.

56 Mr Miles marked on a copy of Exhibit 16, a plan of the area of the fall in which stook X should have remained, the amount of coal left in the last lift before he went to crib. He felt that the scale reconstruction of stook X shown in Exhibit 16 was approximately the form it was taking at that time. Upon inspecting the residual stook X it was clear to him that its square end had been created by the miner. He felt that the representation of the remaining stook X in Exhibit 16 would be close to what he saw.

57 Mr Dunn, the driver of the shuttle car behind the miner at the time of the fall also gave evidence. His memory of the day and the events was very poor. He could not remember sequence 5 in 304 panel on 17 July 1998. He could not remember the lift being undertaken at the time of the roof fall. Nor could he remember any bolting pattern different from normal in the area on the day. He could remember being on the shuttle car in the right hand end of sequence 5 and that the roof conditions on that day were good. He remembered the roof falling on the miner and that he was approximately 2 feet from it. He remembered Mr Gardner standing in the props, in the area shown on Exhibit 16. Of the fall, all he remembered was "just the crack. Just the bang, like a prop breaking, that's it. Nothing else." He also thought that the roof fell while he was on the second load for him after his crib break. He could not recall what coal was left standing.

58 Alyn Rayne Gear had been working at Awaba Colliery for approximately 23 years with another four or five years at other collieries before that.

59 On 17 July 1998 he was operating the miner in sequence 5 starting about four or five metres into the second last lift, the afternoon shift having finished at that point. He said the deputy, Mr Gardner, tested the roof with a 5 foot bolt steel and it sounded quite alright. He cleaned up the stub and intersection of 3 heading and 25 cut-through where loose coal had collected (called "stowage"). This took three or four shuttle car loads. A scrubber box on top of the miner then came off. He brought the miner back to the intersection for repair. He then went to morning tea. Upon his return he replaced Mr Edwards who was driving the continuous miner. He had commenced to mine the lift; Mr Gear took over the miner.

60 He said while he was doing the second last lift ". . . . we had a little bit of flaking at the front of us . . . . in the goaf area itself, just droplets coming down every so often". He identified the area in the middle of sequence 5 at about the fourth and fifth last lift. He said he saw "little flakes coming down every so often . . . . just like a bit of stone dust trickling down at times".

61 Asked whether he took any action as a result of seeing those flakes he said "Yeah, when we seen the flakes, Greg and I just said, why don't we just leave the bit of fender there just to see what happens?". This was a reference to the fenders left in the area opposite 2 heading. The fender was probably 2 metres or less wide. It was about 5 metres into the cut (which was 12 metres leaving a fender of about 7 metres depth or less).

62 In cross-examination he described the props which had been placed to the right of the intersection and also at 25 cut-through opposite the second last lift. The latter showed no signs of weight coming down upon them. He agreed that the flaking roof was similar to that he had experienced in earlier pillar stripping operations.

63 He then described the work undertaken prior to his being relieved by Mr Edwards:

Q. Sorry, you then went off to crib having formed the first part of stook X to about 5m or so into the last lift? A. It would have been a bit longer than that because we stopped at 5m and I went back in and cut another car of coal.

Q. You stopped at about 5m, didn't you, and then you trammed back in -- A. Yes.

Q -- and what you described as a little bit of coal fell off the stook at about that time? A. At about that time, as I was going to go in to do the next car, I cut that piece off.

Q. This was on the goaf side of the part of the stook at the point at 25 cut-through? A. That is correct.

Q. It was that that you estimated to be half a metre to a metre, was it? A. That was half a metre, a metre.

Q. Again, when you're talking half a metre to a metre, are you talking back into the goaf? A. Yeah, you're talking - yeah, into that area.

Q. You continued to mine the extra car or so load of coal? A. That's right.

A. When we came back out, I'd cut another car, but then we came back, Greg Gardner came back in with two props, two or three props - I think it was two props - so I brought the miner back out again. Then I stopped and gave him a hand to put them up and then Barry Edwards came in and relieved me then.

Q. So these were additional props? A. Yeah, we talked about it - when Greg told me to stop at 5m, he said to me, "What do you reckon? Why don't we put up an extra couple of props". We measured them, he went out and cut them. I don't know whether he brought them back in the car or he carried them in. By that time I went and cut another car of coal. As soon as he came back in, I stopped, brought the miner back to the couple of metres back out. Then I went and gave him a hand to put the props up and then Barry Edwards came in.

Q. Where did he put those props up, those additional props? A. Those props went into that part there, into that area here.

Q. Now, I'll describe it for the record. Are you indicating with your finger on exhibit 10 the area in the semicircle around from the left-hand rib of 3 heading across the front of the 25 cut-through on the right-hand side of 3 heading; is that right? A. Yes.

Q. So that the additional props would have the impact of being across in part the stub of 3 heading? A. Yeah, and closer to the intersection itself, yeah, in that area.

64 Mr Gear confirmed the form of the intended stook as conforming with the green area marked on Exhibit 16. He assented to the proposition that the form was dictated by the trying to straighten up the continuous miner.

65 In cross-examination he agreed that stook X would, because of the turning difficulty with the miner, always be wider at the base than the point at which it started. I understood this to be contrary to his evidence in chief that the attempt was to create a stook of a constant width of 1.5 metres. Nothing seems to turn on that variation in his evidence.

66 In re-examination the suggestion that a couple of extra props be put up, referred to in the transcript extracted above, as a question from Mr Gardner to Mr Gear, was said by Mr Gear to have come from him.

67 Also in re-examination Mr Crawshaw sought to have Mr Gear mark on a copy of Exhibit 16, the exhibit showing the allegedly correct formation of stook X, and that which was actually remaining after the roof fall, that part of the correct formation of stook X which was remaining when he left to go to crib and Mr Edwards took over. The transcript reveals the difficulty which Mr Gear had understanding what was asked of him. Nevertheless, he ultimately marked Exhibit 16 in a fashion which demonstrated that the last lift he was undertaking actually cut into the intended correct formation of stook X. Given the difficulty of understanding which was evident, further cross-examination on that matter was permitted. Utilising a scale model continuous miner upon the floor plan set out in Exhibit 16, it became clear that the representation in Exhibit 29, showing the removal of the stook by Mr Gear, could not have occurred.

68 Mr Warren Etheridge was called, he being a coal miner of more than 30 years' experience, all the time at Awaba Colliery. Mr Etheridge had a very limited recollection of the events of 17 July 1998 and nothing which particularly bears upon the matter.

69 Mr Gregory Ian Cornwell, the Electrical Fitter on shift on 17 July 1998, gave evidence. Mr Cornwell had been at Awaba for two to three years, then having had 22 years' experience in the mining industry. Mr Cornwell described the state of the roof conditions in 304 panel inbye of 25 cut-through on that day as very quiet. He developed that answer this way: Well, this is not exactly what you would call a pillar section, where you take out all the coal; you are only taking out strips of it. Normally a lot of time it will stay up in the air and won't fall in, so you won't get a lot of weight on it but the quietness means - timber makes noise, they set timber to the roof, it takes weight, the cracking noise is quite an eerie sound, breaking of timber. Stone will make noise, the roof can crack and bounce and bang. There was none of that. That wasn't any alarming sort of noises or noises that would make you somewhat concerned. (sic)

70 He had recollection of some cracks in the roof in another panel to the left of 304 panel. There were splash faults there as well, but where they were not working.

71 At the time of the accident, Mr Cornwell was driving a shuttle car servicing the miner during the last lift. He was in the right hand shuttle car facing back outbye. He gave this evidence: Q. Could you see what the continuous miner was doing at the production face where you were sitting? A. Certainly.

Q. Had the stook been formed when you were -- A. There was a thin edge running up along the edge of it when I was up there, yes. I don't say it was as wide as what it should have been, but there was definitely a stook that was there, that was part of it, yes.

Q. Well, how wide was it? A. I couldn't tell you. I don't know. I am sitting on this side, this stook is over there. I don't know how thick it was - that thick, this thick, that thick, I'm not too sure (indicates).

Q. You said you don't think it was as wide as what it should have been - do you know how wide it should have been? A. 1.5 metres, I think the rules say.

Q. Was it 1.5 metres wide? A. No.

72 He depicted upon a copy of Exhibit 11 the approximate position of the miner when he was last working the shuttle car before the roof fell. It showed the miner alongside the foreshortened residual stook but with an estimated extension of that foreshortened stook back out towards 25 cut-through, although in a form smaller, as he had earlier indicated, than was planned. He indicated that as far as he could tell stooks were always wedge shaped. He estimated that upon handing his shuttle car over to Mr Dunn to operate it was 10 to 20 minutes before he heard the roof fall.

73 Mr Stephen Cull was the Technical Service Coordinator at Awaba Colliery at which he had been employed between October 1975 and October 1998.The mine deputies were the next rung of management below Mr Cull; the Mine Manager, Peter Foster was the next rung above him. Mr Cull performed the function of undermanager, which duties Mr Foster was carrying out on 17 July 1998.

74 Mr Cull was on leave on 17 July 1998 but had attended the mine on 16 July. He did not then inspect 304 panel but had inspected it on 15 July on afternoon shift. He met mine deputy Johnson on that shift and inspected the left side and decided to abandon sequence one because of conditions they found. The area had been fenced off prior to this. The conglomerate roof was inferior and sounded dull when tested with a drill steel. He said one could visually see that it was not good conglomerate roof. He said: A. You can see, when you look at the conglomerate, the actual constituency of the stone. You can visually, with experience, tell whether it is good conglomerate roof or not by the size of the pebbles in the stone and by where the miner had been cutting. You can pick that.

A. There were, but there were very large pebbles which indicated - and it had already been sounded and it was no-roaded, so it was inferior conglomerate roof.

75 He also said that after inspecting the area with Mr Johnson: A. After I had made the decision that we would flit across to sequence 2 after we finished sequence 5, we went to the face. I walked across between 3 and 2 heading, towards the continuous miner. I picked up a drill steel that was alongside sequence 5 and tested the roof across 25 cut-through to where the continuous miner was lifting. In doing so, a couple of the guys said to me, "Oh, we have already tested that, it's as hard as hell," were their words, "It's hard as hell, there is no problems in 25 cut-through," where I tested. There were no unusual areas of roof that was of any concern.

Q. I am sorry? A. There were no areas - it was all competent across 25 cut-through. There were no areas of unusual nature across there.

HIS HONOUR: Q. Was that told to you, or was that-- A. No, I personally tested the roof myself. I then had discussions with the guys, be them general discussions, as I usually did, in relation to how they were going, the working place, what problems they had, any machinery hiccups that we needed to attend do, and whilst I was at the miner, I did a visual inspection of the area, what would be known as the goaf area, from the continuous miner, and everything appeared, from what I could see, intact.

76 Mr. Cull said the continuous miner operator, Alyn Gear, had mentioned to him that there was some "brat" - a stone band that sat between the coal and the conglomerate roof and they were trimming it down going into the lift and there was a little bit of flaky sandstone although he was not sure it was sandstone.

77 He described the mining process as open-ended lifting which means that coal is not left on the left hand side of the miner whilst it is lifting. However, when they were in sandstone and mudstone roof, "the fender was pocketed so that you would actually have coal on the left-hand side of the machine so that any splashes or bits of sandstone or conglomerate that may drop out were confined to that lift only". He described this as in effect putting in extra stooks as extra support. They were left in. He said that was a matter of practice rather than requirement under the Rules. Alyn Gear indicated to him that this would be done.

78 He said that in the course of taking ventilation readings he had regularly walked along 25 cut-through, passing the area that became sequences four and five in 304 panel. He was asked: Q. What did you notice about the roof conditions in that area? A. Which area?

Q. The area of 25 cut-through and inbye of 25 cut-through? A. I noticed that there were a few spot bolts, which would be additional support installed in 1 heading and inbye of 1 heading 25 cut-through. I noticed that there were four W straps installed in 1 heading, inbye of 25 cut-through, at the stub face. 24 cut-through was clear. 23 cut-through there were - I don't remember seeing any additional support in the other areas. In fact, 304 panel in its entirety was, I recall, all conglomerate roof.

Q. Did you make a note of those observations of W straps in 3 heading and the pattern of bolting in 1 heading? A. No.

Q. What did the W straps indicate to you? A. Oh, it indicated that we had inferior conglomerate roof and that the roof was broken and the W straps had been put in to tie the roof together.

79 He said the additional support in 1 and 3 headings would have been marked on the production deputy's report plans when the place was driven during first workings. That plan was kept at the production deputy's station, which is the crib room and in most cases that was transferred onto plans kept by Mr Cull in his office after discussions with the Deputies when they reported these anomalies or inferior roof.

80 He did not recall whether 1 and 3 headings were marked on any plan "because it was driven a long time prior to us coming back and stripping. However, on my inspection, it wasn't marked on any plans for stripping purposes".

81 The production reports from the first workings were filed away on the surface, either in the main office or the deputy's report room. Twelve months of reports were kept in the deputy's report room and thereafter filed in a filing area.

82 Mr Cull gave the following evidence as to extra support and testing the competency of the roof: Q. Prior to this accident on 17 July 1998, were the workers at the mine required to provide extra support by way of bolting in the vicinity of faults? A. No.

Q. What was the method, or what were the methods, by which you expected workers at the mine to test the stability of the roof prior to the accident? A. To test the competency of the roof?

Q. Yes. A. There were numerous ways. The first one, I guess, is with the continuous miner operator, as he is advancing and cutting, whether he trims the coal to the roof on the advance or on the retreat. He has a fair indication by the reaction of the picks on the roof as to whether the roof is competent conglomerate roof or not. The second method is a visual inspection, and everybody is aware of what conglomerate roof looks like, what sandstone roof looks like and what mudstone roof looks like. The third is a testing method with a drill steel or hammer, the sounding of the roof to test the skin and up to about half a metre into the roof of conglomerate roof. The fourth method is by drilling the roof - test-boring the roof - to a depth of 1.5 metres, checking the finds and the tailings from the drilling, checking for any discolourations which may indicate partings, and checking drill-rotation, because if you do get into any softer-type strata, the drill will jam and rotation will slow up.

Q. We can just go to the last method. You talked about looking at the tailings? What does that process involve? A. Well, the tailings - you use water to drill the roof, to suppress dust. The tailings are a gritty tailing that comes out of the conglomerate roof. If it is a mudstone-type roof it will be a mud tailing that is easily crushed in the fingers. And sandstone may be fine grained or it may be as competent as conglomerate in some cases.

Q. So does that test allow one to know whether there is sandstone or mudstone within what is apparently good conglomerate roof, even though it is not actually visible? A. It forms part of the matrix of the conglomerate, it could be up in the roof somewhere. Mudstone, yes. Certainly sandstone may be a little bit more difficult, but you may get a discolouration in the tailings with different coloured sandstone.

Q. In those circumstances, namely the encountering of mudstone or sandstone that wasn't visible at the face of the roof, if I can use that word-- A. Yes.

Q. --what you would (sic) expect the employees to do? A. Install extra support.

83 He described the effect of sounding the roof with a metal bar: A. Well, you get a vibration or a sounding of the roof with a solid conglomerate roof - it gives a nice sound ring, high-pitched ring. When you have partings, specifically mudstone partings or lenses in the roof, that ring is absorbed by that parting and it gives a dull sound instead of the nice high-pitched ring sound.

Q. I understand that. I am just trying to understand what the relevance of the half metre was? A. Well, okay. Solid conglomerate roof, when you sound it - to get above a half metre is fairly difficult with the sounding method, whereas anything less than half a metre, from my experience, is - you can pick up any inferior roof. Obviously, the skin of the roof is easily picked up, the actual immediate roof, whether it is dull or whether it is sound or not. If there are partings up to half a metre then they will be - that can be revealed through the sounding method.

84 As to a risk assessment on pillar stripping, Mr Cull gave this evidence: Q. Did management at the mine do a risk assessment on the stripping system? A. We developed safe work procedures for extraction of 33-metre centre pillars, and those safe work procedures highlight the steps in lifting and the hazards and the risks associated with those particular tasks that you carry out. As far as --

Q. I am sorry, where was this contained? A. That is a safe work procedure for total extraction for removal of - we did 33- and 36-metre centre pillars. Now, there are safe work procedures for those. What we did was we grabbed one of those fenders out of the total extraction system and developed our pillar stripping system. So the actual system of lifting was still the same.

Q. All I am asking you is whether you had a separate risk assessment-- A. Not a formal risk assessment as such, but safe work procedures were developed for lifting of fenders.

Q. Where were they to be found? A. They were in a procedures manual document kept at the mine.

Q. Who developed them? A. The crews at the mine in control of an undermanager or a deputy. So the actual crews that were doing the lifting developed the safe working procedures, because they knew how to do the job best. It was overseen by obviously myself and the manager, because they became part of the safe working procedures at the mine.

Q. I just want to ask you about the meetings. You say there was a meeting that occurred at the time you changed to pillar stripping? A. Yes.

Q. Who attended that meeting? A. The manager, myself, and the crews on day shift and afternoon shift.

Q. What-- A. Oh, not all together, no, we had a crew meeting on day shift and a crew meeting on afternoon shift.

Q. What occurred at each of those meetings, do you remember? A. There was discussion points brought up about why we stopped the total extraction; why we were implementing a system of pillar stripping, because it was a safe method; the issue that was associated with total extraction was wind blast. The issues associated with pillar stripping were negligible. It seemed to be a safer system of mining that eliminated the problem of total extraction and wind blast.

86 As to leaving stooks he was asked: Q. Were there any meetings at which the idea that you told us of leaving fenders or stooks, where not required by the rules, was discussed? A. This was in mudstone roof where we actually pocketed the fender. They were underground in the panel. I had numerous discussions with the crew. It was the crew that suggested that we leave these fenders alongside the machines so that we isolated any laminates of mudstone that may come it. It was dependent on the conditions they were working in at the time, so the deputy and the miner driver would make that decision. 87 In describing his normal duties in 1998, Mr Cull dealt with the development of sequence plans and the design of pillar extraction panels:

A. Well, I really - the design was, like I have said - it just didn't happen like that in the couple of months leading up to it. At the mine - we'd used it in 5 east, in 7 north-west. The design was open-ended lifting, which had been a practice at the mine ever since I'd been there using continuous miners.

Q. Was there any aspect of the design that took place in the months leading up to July 1998? A. In respect of barriers that were left to segregate the panels, yes. The width of barriers and the size to segregate each individual panel.

Q. Are they reflected on exhibit 10 which is before you? A. No, they are not shown on there. If you had a larger plan of the mine they would be, yes.

Q. There is a plan of the mine on which these barriers are shown? A. They do show, yeah. I can explain to you where they were and what they were there for: there is no barriers actually shown on there as such to segregate that panel, but see these "in" pillars over here, you can see "IN" - there is more to this plan. Those are barriers that were left in so this section was segregated to this section. We don't actually strip or segregate those pillars.

Q. Was there anything else that took place by way of design in the months leading up to July 1998? A. As I have explained in the other production panel, we had - introduced pillar stripping in the barriers. That was a different system than we'd been used to and the reason for implementing that system, which was only a deviation of the total extraction system, was to minimise or to prevent wind blast.

Q. If I could just come back to my question, who is responsible for the barriers? A. Well, they are shown on a plan that is submitted to the department which is signed by the manager. We were guided by - in most instances by external consultants and information that we had been given over the - in the early '90s on workshops that the deputies and I had attended which give us some design principles to work with to be able to put together a panel and extractor.

Q. Can you now answer my question as to who is responsible? A. Well, I don't ultimately know. Ultimately, the manager, but a lot of others had input into it before it got to him. Because he signs off on it - I don't sign off on it. The deputies don't sign off on it. We have a lot of input into the design and, primarily, I would have had 304 panel and 7 north-west, the sequence plan I would have designed.

Q. Were indications of poor roof in 1 heading and 3 heading inbye of 25 cut-through taken into account in determining the lift-off procedure there? A. No.

88 In cross-examination he explained that the external consultants "were used to do geotechnical surveys, to identify various features within the panel". He also said that there were workshops on principles in pillar extraction - "an instrument to help you design pillar panels and to put a system of safe extraction together utilising the principles that we were given". He said those workshops were done externally by Professor Jim Galvin, Ian Anderson and another professor from the university. Mr Anderson was identified as being in court and attached to the Department of Mineral Resources. He stated that it was part of his function to observe that there was compliance with pillar stripping plans as signed off by the manager and that he did observe that compliance.

89 Mr Cull described the way in which events on one shift were conveyed to the incoming deputy either via notes in the diary, or in the case of anything out of the ordinary, by telephone. In relation to the type of roof support applied in 7 north-west panel, Mr Cull stated: A. Conglomerate roof was a test bolt every six metres, which was to bore and test the conglomerate roof. The mudstone roof was a minimum of four bolts in a W strap on I think 1.5-metre centres. If the mudstone, which it generally was, was broken, then we would mesh, which utilised 1.5-metre mesh modules with four bolts in each strap that was part of the mesh module. So on 1.5-metre centres, we had four bolts installed.

Q. Is there any such thing as drummy roof when you're dealing with mudstone? A. No. Mudstone is drummy roof. That is it.

Q. So when you're dealing with mudstone, you just know the roof has to be supported? A. Oh, absolutely, yes.

Q. So you have support rules for its support? A. Yes. Well, it is the top-end trigger of the support, mudstone. It is the densely supported types of roof strata.

90 Mr Cull described it as standard practice to test the roof and if it was found to be other than competent, extra support would be put in. Generally for unbroken non-conglomerate roof butterflies and roof bolts were adequate to support that area.

91 Mr Cull described the partial pillar extraction undertaken in 304 panel as being used because "To us it was a safer method". He said that they had done lifts in 5 east panel "in the barrier where we had had stubs and I'd seen them lifting out through there and forming stooks that were necessary for the intersection protection".

92 He described the shape of stook X, had the pillar extraction rules been complied with, as "not too much different to that shown in the shaded green area" (of Exhibit 16).

93 Dennis Charles Dellaca, a mine deputy employed at the Awaba Mine in 1998, was called in the prosecution case. He worked in 304 panel the day before the accident, the subject of these proceedings. He could recall no unusual roof conditions in the panel prior to the accident. He described the roof this way: "There was conglomerate roof, which was a normal roof. There was - in one part where we worked previously there was a fault. And then there was an area that appeared to be a localised area that had been supported with W straps."

94 He said: ". . . in the area of the fault, small pieces had fallen off in the days previous, which is not unusual in mining."

95 He said "the evidence of fall was not particularly large", large to him meaning where the whole area falls.

96 As to the W-straps in the stub of 3 heading, he had observed those straps. He was also aware of their presence from first workings when he had been involved and possibly in their erection. In describing the roof in that area, he said that he saw what appeared to be non-conglomerate roof "a less substantial structure than conglomerate". It was possibly mudstone or sandstone or a mixture of the two. He said: ". . . conglomerate has got pebbles and stones in it that are very obvious. Non-conglomerate roof usually doesn't have the pebbles and stones. It looks like sandstone. Mudstone can, in some instances, look similar to the eye and you immediately assume it's - that it's non-conglomerate and you treat it warily.

Q. What action did you take after you noticed these straps in the stub of 3 heading in the days prior to the accident, if any? A. Other than doing my daily checks to ascertain whether there was any deterioration of the roof, nothing, really, because the straps made it obvious that there was a difference in the roof structure."

97 Mr Crawshaw, cross-examining by leave, concerning the occurrence of a fall in 304 Panel on the previous shift attended by Mr Dellaca obtained this evidence: Q. You see, the fact of the matter is that you did experience a fall, didn't you? A. Yes, after being refreshed by this report, but you asked me to remember and now I still do not remember the fall. I accept that it must have fell whilst I was there, after having read this, but to ask me to actually remember the fall, and you ask the size and the details - it was, to my mind, insignificant, and I stated that, because it is not unexpected in the area of faults to have small falls of roof material, and so because it was insignificant I didn't make any additional reports about it.

Q. Well, you wouldn't describe a fall that was two lifts wide and about 400mm thick - perhaps thicker in parts - as insignificant, would you? A. In that place, yes, I would, because of the area of the faults it's not insignificant, a fall of that size. It is an expectation that something may happen. (The reference to "not insignificant" in the last answer was later corrected to "not significant").

98 Mr Dellaca was shown a statutory report compiled by him in relation to the shifts on 16 July 1998. He explained that the roof fall in sequence four was not recorded: A. Because it was not a normal procedure and it was not, in my understanding, a normal expectation to record details of falls in areas where - that had previously been worked and that nobody was ever going to go there again because it was, for want of a better term, in a goaf situation and because there was no danger - there was going to be no danger, it probably would never have been - not recorded. And that's the reason why there's nothing on there.

99 In cross-examination the options available to a deputy when bad roof is discovered were described this way: The first option is to fence it off. The second option is to maybe scale it down, if its - if you think it might be just a small thin layer. And the third option is to do something to support it, like putting W straps, mesh, roof bolts. So they are the three options.

100 Mr Dellaca described the way in which the deputy checks the roof every two hours where the continuous miner is working and otherwise on a five hourly basis. His statutory report for 16 July 1998 would have recorded any other drummy roof if it had been found but noted the roof as "secure".

101 Robert George Horn, another mine deputy who had worked in 304 panel on 15 July, but was away on 16 and 17 July 1998 gave evidence of his experience in 304 Panel. He had observed the W-straps in the stub of 3 heading. The W-straps indicated to him that there was an area of drummy roof. He did not mention the straps in any statutory report, as he said he had no reason to do so, the roof had been secured.

102 Paul William Johnson, a deputy for 15½ years at Awaba Colliery up to July 1998. His statutory report for 16 July shows work having been undertaken in sequence four. He observed a fall from the roof in sequence four panel near the fault. The material coming from the roof being approximately 6 inches thick. The fall occurred in a goaf area which had been lifted. It covered an area of 2-3 metres. He recorded this in his report as "sandstone band on roof falling out". He said in evidence that: "I put that down as a warning just to keep an eye on the edge of the lifts. . . . the rock wasn't big enough to stop you mining, but it just - there is a capability when they got to the end of the lift that some might have peeled off and fallen on the end of the miner".

103 Mr Johnson described the circumstances where W-straps were put up where a deficiency was found in the roof. That is, drummy roof or laminated roof. He said: ". . . the drummy sound doesn't tell me what it is. It could mean that there is just a small band up in the conglomerate that - between layers of conglomerate that - you don't know.

Q. What I want to ask you is, in circumstances where you had ascertained there was drummy roof and you had done your drilling and there was no evidence of non-conglomerate roof on the surface of the roof, as it were, so all you had was drummy conglomerate roof as far as you were aware-- A. Yes.

Q --what remedial action did you take in those circumstances? A. The normal practice was, if I had a report it was drummy or I found it was drummy, I'd come back outbye 1 or 2 metres from where I was stopped until I got a ring sound and I'd start actually supporting under the solid roof and then work into the drummy area.

Q. Using what? A. At that time it was a Falcon roof bolter and W straps.

Q. So is this the case: you used W straps for conglomerate roof - drummy conglomerate roof? A. Not at all times. Sometimes we would spot bolt it with butterfly straps, which is - which is a W strap cut into smaller pieces.

Q. Is this the 3 x 2 butterfly bolt pattern you are talking about? A. Yes. Yes.

Q. When did you use the 3 x 2 butterfly bolt pattern and when did you use W straps? A. For me, in charge of the unit, if I thought we were going to be continually wheeling under that road, to give both shuttle cars coverage right across the road we use a W strap. If it was an isolated area we would use butterflies, but at the same time if it was a small drummy patch in the roof we just put a couple of butterflies in that too. They have actually got a metre support pattern on them, so.

Q. Can I ask you this, if you came across drummy roof where there was evidence of non-conglomerate roof either on the surface or through the drilling of the test holes, what remedial action did you take in those circumstances? A. It depends. W straps it if it was - through testing, if we found it is only a thin layer, we might even put mesh up just to stop it. If there was - if I thought that it was going to break up at some time later on, through weathering, I would probably mesh it and that just stops bits falling out.

104 As to the W-straps in 3 heading, Mr Crawshaw asked: Q. Mr Johnson, when you saw the W straps in 3 heading in 1998 did you take any action in relation to that find?

A. There was nothing unusual about finding W straps there. It just meant there was a patch of drummy roof there and I did test up to it, just to make sure it hasn't weathered any further out, and according to my report everything was fine. . . . There was no reason to take action. It had been supported.

105 Mr Johnson also gave this evidence in cross-examination:

Q. I think you also said that you had inserted W straps in other areas of the mine? A. That's correct.

Q. Were any of those areas that you were thinking of when you provided that evidence to the court areas of roof that you observed to be conglomerate roof? A. I'd say yeah, 99 per cent of the time. 99 per cent of the time I got drummy roof it was under conglomerate.

Q. And you observed on occasions such as that that W straps had been put up? A. That was standard practice for us, yes.

Q. It was a practice that you, yourself, followed? A. Certainly.

Q. If I could just take you back, you also told us that when you observed the W straps in 3 heading in July 1998 you tested up to it. Do you recall saying that? A. Yeah.

Q. What did you mean by testing up to it? A. Well, where the roof was already supported with the W straps - a metre apart, or whatever the rules said they were to be apart - I would have tested right across the face line up to the edge of the straps and, keeping in mind even though that roof if it was drummy and it was fully strapped you would still get a drummy sound on that roof anyway. So I'd just test it to make sure that the drumminess, through weathering, through age, hadn't actually crept out into the solid conglomerate which, by my report, it hadn't done, so --."

106 Following the accident, the Department of Mineral Resources commissioned a report from Dr Konrad H R Moelle, Dr.Phil., FAusIMM, MIAEG, MAGS, into the geological and engineering-geological aspects of the roof failure in 304 Panel on 17 July 1998. He provided a report in September 1998 but subsequently passed away. Dr Moelle's report was tendered in evidence and became the subject of explanatory comment by Mr Darryl Stevenson, the principal of Geological and Mining Services Australia Pty Ltd., who effectively concurred with the conclusions arrived at by Mr Moelle, given Mr Stevenson's knowledge of the Great Northern Seam and the Awaba Colliery.

107 Dr Moelle's conclusion was, in part, that: "The failure in 304 Panel is most probably attributable to a combination of an infrequent depositional configuration and lithological change in the immediate roof (5m to 8m), and the current layout of the 304 Panel. The failure seems to have occurred when a conglomerate unit parted from an overlying laminite succession that may have formed in an oval and oblong depression in the conglomerate immediately overlying the Great Northern Seam. Part of the laminite succession also failed.

In the present author's opinion, it is extremely difficult to identify areas which are affected by significant depositional irregularities in the immediate roofstrata and the detailed mapping that would be required to identify such areas is probably a requirement that is unrealistic with regard to the likely costs involved in conducting such exploration programmes. It is, however, recommended to use the existing database built on drilling data to establish a depositional model and a roof rock distribution map based on bore information. Such a model could possibly be refined through digital manipulation, and could conceivably reach a state of refinement that would indicate areas where particular caution is required.

It may also be helpful to do a very detailed mapping programme in the 304 Panel area close to the accident site, in order to identify characteristic features caused by differential compaction that are indicative of significant lithological changes in the immediate roof. For sites like the failure described above in 304 Panel, it is recommended to rely prominently on remotely controlled equipment, as failures of the type under consideration here will always occur without sufficient warning. In the opinion of the present author, the failure has been caused by geological responses of a complex sedimentary succession to mining induced loads.

Geophysically based remote sensing, combined with digital modelling of the palaeoenvironment at Teralba Conglomerate time, will probably improve the operator's ability to recognise those areas which have particular and unusual support requirements. It is also recommended to initiate a government controlled and funded research project into the requirements of so-called strong roof strata, as a matter of priority.

108 Inspector Beatty also agreed in evidence that he drew a conclusion that the detection of the flawed roof structure prior to mining was difficult, if not impossible, with the limited resources, equipment and expertise available as at July 1998.

109 Kelvin Joseph Ryan was the first witness called in the defence case. Mr Ryan was the Company Secretary of Powercoal. His evidence related to the management structure and the means by which Powercoal related to Pacific Power, the holding company of Powercoal. In particular, it dealt with the limits on the responsibility of mine managers in the context of management of Powercoal. This material was relevant in the context of the charges brought against Mr Foster, as a person "concerned in the management" of Powercoal. For reasons which will become clear, it is unnecessary that I detail these matters.

110 Dr James Maurice Galvin gave evidence. Dr Galvin is a Professor of Mining Engineering at University of New South Wales but now is on sabbatical leave. He described the test method employed in the industry when working under conglomerate roof in first workings as being: 1. The continuous miner cutter heads will make a sound on the roof which may actually be felt. 2. Visual inspection and then testing by steel. The miners will work their way in progressively by sounding the roof. They will install support and, if concerned, may install more support. When drilling, they will observe the colour of the cuttings, the noise of the machine, and whether the machine is holding or jumping while cutting. In Dr Galvin's opinion this was the appropriate method in 1998. There was no other proven system. There was no other method to test the roof other than with drill steel or hammer.

111 In cross-examination, an attempt was made to discredit Dr Galvin to a degree questioning his independence. However, he later explained that the Department of Mineral Resources was seeking to have him prepare a report on a brief of evidence concerning occupational health and safety liability in a south coast mine, not in the coal industry. He was considering a question of conflict at the time of giving his evidence. He is an adviser to a variety of government departments. I am satisfied on the evidence as a whole that there is no proper basis for doubting the independence of Dr Galvin in expressing the opinions on the matters raised with him relevant to these proceedings and more generally. I am satisfied that he attempted to give his evidence as a person who had formed an opinion of the events and circumstances based on his acknowledged expertise.

112 Dr Galvin prepared an extensive report into the incident of the roof fall on 17 July 1998. He made three visits to the accident site in August, September and October 1998. He made a series of observations and conclusions including the following: 1. The stress environment in the area was benign.

The fall basically consisted of a wedge of conglomerate overlain with laminite that cantilevered off the pillars on the outbye side of 25 cut-through .... the fall occurred when the bending forces generated by the cantilever exceeded the tensile strength of the conglomerate close to the 25 cut-through abutment.

The conglomerate that fell contained a number of sandstone lenses up to 500mm in thickness. The sandstone appeared competent and hard and the lenses were "welded" into the conglomerate so as to be very unlikely to give rise to partings . . . competent sandstone lenses . . . are not readily detectible by drilling and sounding.

The fall could not be attributed to the presence of joints. However, the outline of the fall at 25 cut-through, 3 heading intersection was determined by the NW-SE joint set . . . .

The NW-SE joint was very tightly closed. Joints that could be detected on the face of the fall could not be traced along the roof of the workings that had not fallen. They could not have been mapped prior to the fall.

The immediate conglomerate roof appeared visually to be consistent in composition and to be competent in its own right. That is, it did not display signs of being composed of different sorts of stone/matrix (aggregate/cement) mixes or of cracking or fretting (weathering).

113 Dr Galvin also made some general observations of relevance: 1. It is estimated that the Australian Underground Coal Mining Industry installs 5 to 6 million roof bolts per annum. It is unrealistic to think that the geotechnical features which caused this many bolts to be installed are all known. Rather, the reality is that bolts and straps are an engineered solution for working in an environment where there are many unknowns. Provided they are installed in a manner appropriate for the mining system being utilised and that the area is monitored for timely signs of any further deterioration, there is no need to identify the geotechnical features causing their installation. 2. In relation to the accident site, Dr Galvin would not have interpreted the feature correctly on the basis of what he could have observed prior to the goaf fall. At the time, there was no way of determining the nature and extent of the laminite feature from the presence of the straps and bolts. He concurred with the observations of Professor Moelle, who described the feature as an "unusual depositional configuration" and that "it is extremely difficult to identify areas which are affected by significant depositional irregularities in the immediate roofstrata . . . It is extremely difficult to see how any form of conventional roof support could have prevented this particular failure considering the geological configuration, which could most probably not been anticipated on the available geological evidence".

114 He also observed that "apart from the small amount of laminite in the face of 3 heading, all the immediate roof was conglomerate. It would have been appropriate to have mapped it as such".

115 He concluded that given the frequency, magnitude and presence of faulting in the area, on the basis of the information available at the time, he would have associated the roof bolting and the observations of roof fretting and splashes in the goaf with a presence of faulting, based primarily on his practical experience around faults.

116 Dr Galvin noted that Awaba Colliery had practised total pillar extraction in the Great Northern Seam for many decades up until 1997. During this time, serious problems were encountered on a number of occasions with lack of control over caving of the massive conglomerate roof strata associated with this mining method. In September 1996, total pillar extraction was approved for 5 East Panel at Awaba Colliery, which was based on past experience at the mine. The approved plan provided for a tapered stook X with a minimum width of 1.5m to be left as operations retreated into an intersection and also as operations retreated back from an intersection. After commencing the extraction of 5 East Panel in late 1996, by early February four rows of pillars had been extracted without a major goaf fall occurring and attempts to induce caving by blasting from the surface were only marginally successful. Given these problems, it was decided to leave a row of pillars intact and revert to partial extraction for the remainder of 5 East panel.

117 The report stated: I was involved in the design of the pillars in 5 East Panel. Another consultant, Dr J Shepherd, was engaged to design the roof support system. The pillars that remained after partial extraction operations were designed to have a high factor of safety against failure. This was in order to minimize the potential for a pillar collapse or creep overrunning the working area and/or for a major windblast to occur. The system was planned with the intention of preventing goafing on a regional scale; however, the mining plan was based on the assumption that local roof falls may still occur from time to time.

In April 1998 I reviewed various layouts proposed by colliery management for pillar stripping in 304 Panel. By this time, the colliery had developed its own in-house capability to design coal pillars as a result of key personnel attending a series of pillar design workshops conducted by Prof Grant Hocking, Mr Ian Anderson and myself. I validated the colliery's computational design program with Mr Enright and discussed the various pillar size options with management (Surveyor Phil Enright, Undermanager in Charge Steven Cull and Mine Manager Campbell Johnson). (Dr Galvin said in oral evidence the reference to the Mine Manager might be wrong. It could have been Peter Foster.)

118 Dr Galvin drew the following conclusions: The mining layout, the dimensions of the working area and the sequence of extraction in 304 Panel were designed to provide protection to operators from goaf falls and associated hazards such as windblasts and feather edging. The concept of pillar stripping had already been proven in these regards under similar geological circumstances at Cooranbong Colliery, immediately adjacent to Awaba Colliery.

The mining system was based conservatively on dimensions and mining practices that had been used successfully over many years in the more hostile total pillar extraction panels at Awaba Colliery. A comparison between Figure 1 and Exhibit 7 shows that, but for the length of individual lifts, the system of lifting off was effectively the same. Although 304 Panel was subject to lower goaf edge loadings than experienced in total extraction layouts (explained in Section 5.4), nevertheless a conservative approach was adopted and the number and arrangement of breaker props, the angle of lifting and the size and location of stook X were left unchanged.

Mr Anderson has stated that he considers that the presence of four straps and bolts at the inbye end of 3 Heading should have caused management to increase the size of stook X. Based on theoretical and practical considerations I do not concur with this viewpoint. Reasons for this include:

At the time, there was no way of determining the nature and extent of the laminite feature from the presence of the straps and bolts. I would not have interpreted the feature correctly on the basis of what I could have observed prior to the goaf fall. The concerns that Prof Moelle (Exhibit 31) had in regards to identifying this type of feature have already been noted in Section 3.2. Given that the roof outbye of the feature tested satisfactorily, that the system of mining was designed so that no one was near the bolted area during pillar stripping and that the system of mining was designed to prevent the goaf entering the working place, I think that it is highly unlikely that a mine official would have taken the presence of bolts and straps as a trigger to leave a larger stook. I would not have.

From a practical perspective, the stook size had already proven acceptable in the much more hostile environment of total pillar extraction. There was no experience base or reason to suggest that it would need to be bigger.

The theoretical consideration of caving mechanics undertaken in this section of the report provides no reasons for increasing stook size.

Mr Anderson's concept that the stook needed to be larger to cope with dynamic load does not concur with theoretical expectations that the stook would be able to withstand a higher dynamic load than static load. This is because the strength of coal is time dependent (it loses strength over time). As far as I am aware, there was no experience base at Awaba Colliery to suggest that stooks were weaker under dynamic load.

Even after the event, I do not concur with Mr Anderson's opinion. The performance of stook X cannot be assessed because, for all practical purposes, it was not present. I am unaware of any other experiences at Awaba Colliery which would indicate that the design stook size should be increased when bolts and straps are present or that it needed to be in order to accommodate the fall of goaf in 304 Panel.

Obvious control measures when extracting pillars in the vicinity of intersections are to restrict the effective spans and to install support that resists roof displacement in the general area. The most effective means of achieving both of these objectives is through the leaving of stook X.

. . .. the leaving of stook X is one of the most basic elements in partial pillar extraction and in total pillar extraction.

119 Dr Galvin also concluded that: "the probability that the fall would have extended into the intersection of 25 c/t and 3 heading if stook X had been formed as designed is extremely low . . .

However, as Mr Anderson noted in his evidence, the weight of the fall was not distributed centrally over the stook and so this would set up a bending moment (lever arm) that would magnify the load acting on stook X. The magnitude of this load cannot be determined because, due to the discrete cantilever mechanism operating in the immediate roof, a large proportion of the load would have been transferred to the 25 c/t abutment. However, this is largely irrelevant since the main function of the stook in this particular circumstance would have been to break off the fall before it entered the intersection. Given the width, load carrying capacity and stiffness of the stook I have little doubt that it would have been effective in breaking off what, by mining terms, was a very shallow goaf fall before it entered the intersection. Similar sized stooks had performed adequately in much more hostile environments at the mine. In the case of 304 Panel, I have a second reason for believing that stook X would have been effective in stopping the fall before it entered the intersection. This relates to the control that it would have had over the jointing in the roof. Although I do not consider that the jointing played a role in the development of the fall, the NW-SE joint set did define the break off point of the fall in the 3 hdg, 25 c/t intersection. The joint set coincided with the natural path that the fall would take in the corner of the workings. Had stook X been left as planned and had the fall still occurred, it is most probable that it would have terminated along the NW-SE joints on the goaf side of stook X (mirroring what happened when the fall hit the solid pillar at the northern end of the goaf, this pillar effectively acting as stook X in the absence of the designed stook X). (The reference to the solid pillar at the northern end of the goaf in the final parenthesis in this quote is a reference to the right side of the stub of 3 heading).

120 The report also observed that between 1993 and 1998 Dr Galvin conducted five training sessions for various members of Awaba Colliery workforce. The first pertained to conditions of approval imposed by the Chief Inspector of Coal Mines in regards to extracting pillars in the south-west area of Awaba Colliery, incorporating 304 Panel. Condition 4 stated that "for all areas under 30m of cover a geotechnical investigation shall be undertaken to identify possible instability that may influence goaf formation in an uncontrolled manner". Condition 5(e) required "education and training of the workforce in sudden collapse potential". Dr Galvin stated that Dr Bruce Hebblewhite and Dr John Shepherd undertook the geotechnical investigation and he undertook the training. His understanding was that all employees of Awaba Colliery, including trades and management, attended the training sessions he conducted. The training sessions placed particular emphasis on the need to comply with the system of work. Case studies were presented which focussed, amongst other things, on the role of stooks. His report stated: Various colliery personnel were present at three, two day workshops that Prof Grant Hocking (Prof of Engineering Geology), Mr Ian Anderson and myself conducted between May 1994 and July 1995. The first day of these workshops was concerned with pillar design. The second day was concerned with pillar extraction. A strong emphasis was placed on field performance (Ian Anderson), the role and design of stook X (Jim Galvin) and the pillar extraction manual (Ian Anderson). A stook design exercise was included in the lecture notes and undertaken in class.

121 Dr Galvin's records confirmed that Messrs Gear, Dunn, Miles, Edwards and Gardner had attended training programmes, one programme in the case of the miners and two for Mr Gardner, the deputy. Mr Cull attended three programmes and Mr Foster one programme.

122 Dr Galvin's conclusions regarding the accident were as follows: 1. Pillar stripping is a partial pillar extraction method that is based on the splitting and lifting technique used to totally extract coal pillars.

2. Pillar stripping was introduced at Awaba Colliery as a proactive measure to eliminate risk associated with high abutment stresses (goaf edge pressures), violent windblasts, floor heave and feather edges due to irregular and unpredictable caving of the conglomerate roof at Awaba Colliery. This was achieved at the expense of percentage extraction in areas already approved for total pillar extraction. It was achieved by restricting the width of the goaf (extraction voids) and separating the goaves with pillars of coal designed to be stable in the long term.

3. The Awaba Colliery workforce, neighbouring mines and geotechnical consultants were all involved in assessing the pillar stripping technique and introducing it into Awaba Colliery.

4. The pillar stripping system introduced at Awaba Colliery was based conservatively on dimensions and mining practices that had been used successfully over many years in the more hostile total pillar extraction panels at Awaba Colliery. Despite the much lower stress environment and the elimination of full caving to the surface in pillar stripping, a conservative approach was adopted and the number and arrangement of breaker props, the angle of lifting and the size and location of stook X were left unchanged.

Pillar stripping was not intended or expected to eliminate goaf falls. This objective is not technically or practically feasible. The system does not permit access to the goaf in order to assess rock quality and to install roof support. Even if it did, it would not be economic to install the type and density of roof support required to prevent all failures.

Instead, the system is designed to accommodate falls of goaf safely by: a. Being undertaken on the retreat.

b. Setting breaker props at all access points to the goaf, other than the roadway being used to wheel coal, before lifting operations start to impact on the intersection.

c. Leaving a substantial stook X when retreating through an intersection.

d. Immediately timbering off all access points to the goaf after the last designated lift has been taken off a fender and the continuous miner retreated back through the intersection. 7. These controls were all incorporated into the Manner and Sequence of Lifting Plan and the Support Rules for pillar stripping in 304 Panel (Reference for example, Exhibits 7 and 10).

8. All these controls, that is 6(a) to 6(d), were breached in 304 Panel on the 17/7/98: a. Stook X was not formed as designed. For all practical intents, no stook X was left.

b. The line of breaker props which should have been set at the access point to 3 heading stub was compromised after the completion of the designed last lift. Props were removed to permit the continuous miner to gain access to 3 heading stub after completing the last designed lift in sequence 5.

c. The continuous miner effectively advanced out into the goaf. It had no protection from solid coal to its immediate right (because 3 heading was present), no protection to its immediate left (because stook X was not left) and no protection to its immediate rear (because the intersection was present).

d. The continuous miner was not withdrawn immediately after the last designed lift had been extracted. 9. It should not have mattered that the face of 3 heading was bolted and strapped. The roof outbye of the bolts and straps was tested regularly and often during the pillar stripping operation. It was determined to be competent to a standard and extent that past experience under conglomerate roof in the Great Northern Seam indicated should give no cause for concern in the circumstances associated with the planned system of work.

10. It is not a simple task to determine the type of features that give rise to the need to roof bolt in first workings. It is extremely complex and expensive, to the point of being technically and economically unfeasible, to determine the spatial distribution and mechanical properties of features above future goaf areas. They may be present without any indicators being apparent in first workings.

11. It should not matter that features such as that at the face of 3 heading cannot be or are not identified. In first workings, the installation and monitoring of roof bolts is an adequate control for safely managing these features under most circumstances. More detailed investigations and intensive support measures are usually only undertaken after monitoring reveals that the initial support measures are inadequate. There are no indications that the bolts and straps in 3 heading were inadequate for supporting the heading. On the contrary, the manner in which the roof fell at the site would suggest that they were adequate. In secondary extraction, the features are managed safely through the implementation of the controls such as extracting coal on the retreat, erecting breaker props and leaving a stook X. These were incorporated into the designed system of work for pillar stripping in 304 Panel.

12. It should not have mattered that some flaking and brat were evident in the goaf. The system of work was designed with the expectation that falls of goaf may occur from time to time and, therefore, it incorporated adequate controls to safely manage these circumstances.

13. In my opinion: a. The accident occurred because the designated system of work was not executed as planned on the day of the accident.

b. The system of work was not complied with because stook X was not formed as designed and because the manner and sequence of lifting off was not complied with.

c. Had stook X been left, there is a high probability that the goaf fall would not have occurred until operations had retreated out of sequence 5, if indeed, it did occur at all.

d. Had stook X been left, there is a very high probability that the goaf fall would have been broken off by this stook and so not extended into the intersection and 3 heading stub.

e. Had the designated system of work been executed as planned, operations would have already retreated out of the area at the time that the goaf fall occurred.

f. The deviation from the planned system of work occurred over a time span of only some 30 to 50 minutes.

g. The production of a geotechnical map of the area prior to the accident would not have caused the manner or system of work to be changed as the necessary controls were already imbedded in the system.

123 In cross-examination in relation to these conclusions, Dr Galvin adhered to the view that the system of mining was designed to cater for the circumstances which were faced at 3 heading including the bolted and strapped roof. He said the area outbye that was tested and tested again and again. This testing was by sound. He was asked whether that was fairly limited to which he answered: "Yes, but in these circumstances that's what the experience base - the system of mining is based on. That's the knowledge base that we build the systems around, to cater for that."

124 He also said that flaking and brat should not have mattered at all. He said: ". . . it was a goaf and even with a conglomerate roof you can expect some brat and localised pieces of roof to fall down. Because rock has natural weaknesses in it . . . if you open up a goaf area sooner or later something is going to fall. You are going to hit a defect somewhere and it is going to fall. Now, the system was designed to accommodate that.

Q. But if what was falling was non-conglomerate in nature, that was a problem for safety, wasn't it? A. Why?

Q. It was an indication that you had non-conglomerate roof? A. But so what? I mean, there's an enormous amount of pillar extraction that takes place under non-conglomerate roof - probably more pillar extraction has occurred under non-conglomerate roof over the years than under conglomerate roof.

Q. But you don't know how far that non-conglomerate roof extended, do you? A. At the end of the day it doesn't matter, because the system was designed to tolerate goaf falls and, if anything as I said, because the roof is weaker, you would expect it to cave - if it was going to fall, it would fall more regularly, it would cave more steeply, it would snap off easier. So you wouldn't expect it to run into the working place. The feather edging, for example, that is associated with Awaba, you don't get that sort of behaviour under weak roof.

125 Dr Galvin was then asked about his view that the production of a geotechnical map of the area prior to the accident would not have caused the manner or system of work to be changed and it was suggested to him "See, you just don't know that, do you?" He explained that having looked at the area he considered whether he would have called it non-conglomerate roof on a plan and he took the view that he would not have done so. He would have marked the straps "because that's obvious, but I wouldn't have read that feature". He was then asked:

Q. You wouldn't have called it non-conglomerate roof even if it was non-conglomerate, would you? A. I don't think I would have called it non-conglomerate.

Q. Even though it was? A. I could have called it anything. But I would have certainly flagged that there was feature there, but I wouldn't have - I don't think I would have called it non-conglomerate. I would have flagged a feature.

Q. Even if it was non-conglomerate, you wouldn't have called it non-conglomerate; is that right? A. I don't think so, no.

Q. See you'd say anything, wouldn't you? A. No, not at all.

Q. You've become an advocate in the cause of Powercoal Pty Limited, haven't you, in this case? A. That's incorrect.

Q. See, you have been here when Mr Cull has told us that he abandoned sequence 1 because of poor roof? A. He abandoned it because of poor conglomerate roof, yes.

Q. So how do you know that if he'd been fully conscious of the poor roof in the area that we are talking about between 1 and 3 heading that he wouldn't have abandoned that? A. Well, you need to have a look at the circumstances Mr Cull abandoned in. The circumstances were that he was in a heading, still developing the heading, not the goaf, still the heading, and he was on a fault plane and he abandoned it because he was on a fault plane in a heading knowing he still had to drive the heading all the way up, have that feature behind him and then lift off and retreat through that feature with that behind. Now, you just don't do that. Whereas in this case it's quite different because the feature was in front, no-one was going back in there, the mining system was designed to retreat back, so the feature shouldn't have mattered. They're quite different circumstances.

126 It was immediately after this evidence that Dr Galvin revealed the fact that the Department of Mineral Resources was continuing to seek out his services as a consultant and adviser.

Conclusions 127 As to the issue of the nature of the roof under which the miners were working on and prior to 17 July 1998, it seems to me the answer is provided by Dr Moelle's report. He concluded that: It is suggested, on the basis of the available evidence, that the recent roof failure in 304 Panel is largely attributable to an unusual depositional configuration, involving conglomerate and laminite units with extremely weak contacts and very little effective bonding in the immediate roof, approximately 5 to 8 metres above the working roof. The roofbolts used were not able to prevent the failure at this specific site, and it is suggested that other conventional support measures would probably also have failed to secure this particular roof section.

128 While senior counsel for the prosecutor sought to describe the roof as non-conglomerate roof and was somewhat critical of Dr Galvin's evidence in that regard, Dr Moelle, whilst accepting in the quotation above the existence of laminite units with weak contacts and little effective bonding, later in his report referred to the roof as "The failed conglomerate roof segment . . .". Yet it appears to me the term used to describe the roof was not a critical matter.

129 There is in evidence the results of test bore holes drilled prior to first workings which, in the area of 304 Panel, suggest that the roof was sandstone; other like material records it as conglomerate and sandstone. The relevant difference between conglomerate and sandstone roof is reflected in the support rules for 304 Panel, which were approved in May 1998. The bolting pattern required in a sandstone roof is more concentrated than for conglomerate roof. Here the stub of 3 heading was bolted with W-straps, a method not consistent with the more limited bolting design for sound conglomerate roof and was indicative of some different roof condition, but one which was not, on the basis of Dr Moelle's report, the feature which led to the collapse of the roof, given the inability of the bolting process to discover the weakness which caused the collapse.

130 There was also evidence from supervisors and at least one member of the mining crew of their knowledge of the W-straps placed in 3 heading and what that indicated. Messrs Cull, Horn, Dellaca and Cornwell spoke of that knowledge.

131 The prosecution advanced its case in relation to both the assessment and recording and notification charges in the context of a risk of roof fall reflected in the signs of poor roof which were evident at various points throughout the mining of 304 Panel, 25 cut-through over the period 15 to 17 July 1998. Those signs, it was contended, were evident in 1 heading and throughout the pillar stripping operation between 1 and 3 headings and also in 3 heading itself. The recording and notification charge relates essentially to the failure to give prior indication thereby of the discovery in first workings of the poor roof in 3 heading.

132 This approach does not depend upon the skimming of stook X, a concentration upon which it was submitted is misconceived; as wrongly concentrating on the accident itself rather than the risk to safety, which need only be a potential risk. It was submitted that the prosecution has run its case "in a way which makes it possible to find that the charges are made out in relation to potential risk without finding a causal connection with a more limited risk that caused the accident" - as occurred in WorkCover Authority of New South Wales v Howard & Sons Fireworks Spectaculars Pty Limited per Boland J [2000] NSWIRComm 211, 31 October 2000, unreported. An alternative submission was also made, to which I shall come later.

133 There seems to me to be a fundamental difficulty with this approach on the facts of this matter. That there were signs of poor roof is clear. What is not so clear is that there was a risk associated with that roof beyond the ever present and normal risk which the mining system is intended to manage. The minimum support rules provide for testing the roof and, where it is found to be poor, supporting it in an appropriate fashion; the poorer the roof the heavier the concentration of bolting and the application of W-straps and the like. While miners are testing the roof, by employing the recognised means, there is in that process no inherent failure by the employer to ensure the safety of employees. If that were not so, it seems to me the very process of mining underground by these accepted means would have to be regarded as inherently dangerous to the point where the employer has failed, or perhaps cannot but fail, in its statutory duty. That view cannot be supported.

134 In a pillar stripping operation the technique is to mine in a retreating fashion so that the continuous miner is constantly alongside a wall of coal which operates as support for the roof, which is becoming the goaf. That process, which it must be recalled is an approved one, creates a roof which is intended to fall after mining has ceased.

135 In the present case, there is, in my view, no evidence on which it could reasonably be thought that the testing of the roof during the mining process, as it was undertaken in accordance with the mining plan, exposed employees to a risk of a type which could constitute a breach of s15 of the OHS Act. The signs of poor roof were insufficient to cause the miners any concern that it was outside their experience and ability to control; they installed, in a quite normal way, some extra props and left fenders of coal as the means of dealing with the circumstances they were then experiencing. It is appropriate to recall Mr Cornwell's evidence (set out at par. [69] hereof): "Stone will make noise, the roof can crack and bounce and bang. There was none of that. That (sic) wasn't any alarming sort of noises or noises that would make you somewhat concerned".

136 The difficulty here was in part manifested by the unknown horizontal fault in the roof strata which Dr Moelle concluded ". . . could most probably not have been anticipated on the available geological evidence". Mr Stevenson notes in his commentary that even less evidence was available before the accident.

137 Dr Moelle also concluded that, in his opinion, ". . . the failure has been caused by geological responses of a complex sedimentary succession to mining induced loads".

138 Mr Anderson drew the conclusion that the laminite layer in the roof was "almost certainly detected" in first workings and that explained the use of W-straps in 3 heading (see par [32]) hereof). However, I understand that conclusion to be based upon the existence of the W-straps in 3 heading and thus to be an inference which may be correct but equally may be erroneous. Given that the conglomerate layer was too thick for the roof bolts to secure it (at least once stook X had been largely removed) nor for the drills to reach the laminite layer so as to identify its existence, the W-strapping may be simply evidence that the mining crew at first workings discovered drummy roof. Mr Johnson's evidence (see pars. [103] to [105] hereof ) is supportive of this.

139 I find on the evidence of the three experts, Dr Moelle, Mr Stevenson and Dr Galvin, that the actual weakness in the roof which fell was not detectable prior to the fall and that there was accordingly no known or identifiable risk arising from that weakness.

140 The question of the provision of an adequate system for testing the safety of the roof may also be examined from the perspective that the means of testing employed, through essentially sight, sound, and vibration, were recognised in the evidence of the experts as the only means of testing the stability of the roof available at the time of this accident. They were the known and accepted means. In these circumstances, there has not been demonstrated, on the evidence, any failure by Powercoal to provide an adequate system for assessing the safety of the roof in the context of its mining plans.

141 However, at the point of time of the accident to Mr Edwards, the roof was being subjected to loads which were unplanned, so far as the defendant was concerned, by the mining of stook X. The prosecutor also made an alternative submission that the relevant risk was the risk of roof fall and injuries, specific to Mr Edwards' location at the time of the accident. It was submitted that no witness gave evidence that there was no risk of the fall extending into 3 heading if the stook had been properly formed. Further, the duty to provide a risk free work environment is also owed to the "hasty, careless, inadvertent, inattentive, unreasonable or disobedient employee" (per Walton and Boland JJ in WorkCover Authority of New South Wales v Arbor Products International (Australia) Pty Limited (2001) 105 IR 81 at 99).

142 The defence submitted that the matter discloses a demonstrated failure by an employee or employees to adhere to a well-established and well-settled safe work method which has been promulgated to all relevant employees. In addition, there exists a statutory system which confers powers, rights and obligations on those employees. It was the un-notified, unexpected departure by the employees from that system which caused the risk - the unauthorised mining of stook X. Wooden props had been set around the intersection so as to close off the stub of 3 heading. These were removed in order to get the continuous miner into the stub of 3 heading to complete the last part of the mining of the stook. That decision to remove those props was contrary to the pillar extraction plan and procedure.

143 It was also submitted that while Mr Gardner, the deputy, was not able to be called to give evidence in the proceedings, he was placed by Mr Dunn in evidence as standing, at the time of the fall, in the intersection of 25 cut-through and 3 heading, amongst some timber props which had formerly been placed there. This was not a case where it could be concluded that the defendant had failed to attend to risks to the safety of employees and cast the burden on a selected employee such as the deputy. The deputy and the miners themselves were part of an integrated statutory scheme which is designed to ensure, so far as possible, the undertaking of mining in a manner which is safe.

144 It was submitted that the judgment in Arbor Products may be distinguished as a s 18 case relating to supply of equipment; that the suggestion of a duty to a "disobedient" employee is incorrect, but, in the alternative it is a duty only applicable in the case of conduct which is "reasonably foreseeable".

145 There is sufficient evidence to support a conclusion, on the balance of probabilities, that the collapse of the roof occurred as a result of the induced loads imposed on the roof by the mining out of stook X. Both Mr Anderson and Dr Galvin accepted, or described, the purpose of the stook as being to cause a break-off point for any fall in the goaf. That the roof fall broke off around the residual portion of stook X and then into the stub of 3 heading, before breaking off there, seems most likely to have been the result of the virtual elimination of stook X.

146 I am not satisfied beyond reasonable doubt that the prosecution has made out its case on the assessment charge.

147 The faulty roof in the area in which Mr Edwards was working at the time of the fall was indicated by the strapping in the stub of 3 heading. The presence of the strapping itself did not identify any more than that the roof had been found at first workings to require extra support. However, the support of the roof in the stub in that way meant that there was no barrier to persons undertaking work under that supported roof. In its designed state as a stub heading, it was not part of any goaf area. As I have said, that the fall occurred in that area seems to me, on any view of the evidence, attributable to the reduction in size of stook X. The continuous miner was, at the time of the fall, positioned partly in 3 heading and partly in an area which should have been part of stook X. It was actually engaged in skimming the stook at the time of the fall. Stook X was thereby brought to its final, reduced size.

148 It must be emphasized that the charges involved in this matter are not related to the reduction in size of stook X. The failures alleged in the charges go to the absence of any recording and thus notification of the weakness identified in the 3 heading stub at the time of first workings on plans to be used for second workings and secondly, the failure to provide an adequate system for assessing the safety of the roof. As to the recording and notification charge, the prosecution has established that there was an absence of recording on any plans used for second workings of the poor roof condition represented by the W straps in the stub of 3 heading. It follows that there was no notification of that earlier discovery through any plan. In circumstances where the mining officials and crew were aware of the indicia of poor roof in the form of the W-straps, and where the nature of the difficulty in the roof was undetectable at first workings, there cannot be any causal connection between the absence of any recording and notification of first workings information and the fall of the roof or the relevant risk to safety to which the workers, and particularly Mr Edwards, were exposed. The mining plan for stripping the relevant area in 304 Panel catered, whether by deliberate intention or not, and whether or not directly or indirectly, for that area of poor roof by providing stook X in its planned size.

149 While Mr Anderson suggested that the stooks in 304 Panel should have been designed in a more conservative fashion because of the poor roof condition, there is no ability on the evidence to determine whether that amounted to a relevant failure by the employer given the fact that stook X was reduced in size by unauthorised mining to 20% of its planned size. In the developing circumstances the planned size of stook X, whether large enough or not, became immaterial. There is, in my view, no warrant to conclude that the designed size was deficient such that it contributed in any degree to any risk to employees.

150 I am also not persuaded that the notification of bore hole indications of defects in the roof would have assisted with the avoidance of any relevant risk to the crew. On the evidence, any abnormal risk associated with mining was non-existent in the areas in which the men were working until stook X was drastically reduced. Dribbly or flaky roof occurring in the area of the goaf was a normal experience. The continuous miner was retreating from those areas, as intended. To treat awareness of bore hole indications as of any assistance would be tantamount to suggesting that the mining crew would not have reduced the size of stook X, in breach of the rules, had they known of those indications. The roof fall itself occurred while the miner was in a position that was never intended for it, widening 3 heading without the requisite special approval. Yet, had stook X had been as designed, 3 heading was a safe work area with a roof which had been secured. The real point is that stook X should not have been skimmed, but the charges did not go to that.

151 What is not explained in the evidence is why stook X was to be reduced to a point which exposed the poor roof in the stub of 3 heading to the open area which had previously been formed as the goaf, which was supposed to be "protected" by stook X. There was no witness who actually authorised (for example, deputy Gardner) or undertook, or admitted to assisting or any knowledge of, any skimming of the stook.

152 In its decision in WorkCover v Arbor Products, the majority of the Full Bench said: 45 The duty to provide a risk free work environment is a duty owed not only to the careful and observant employee but also to the hasty, careless, inadvertent, inattentive, unreasonable or disobedient employee in respect of conduct that is reasonably foreseeable: Dunlop Rubber Australia Ltd v Buckley (1952) 87 CLR 313 at 320; McLean v Tedman (1984) 155 CLR 306 at 311 - 312 and WorkCover Authority of NSW (Inspector Twynam-Perkins) v Maine Lighting Pty Limited (1995) 100 IR 248.

153 The reference to the disobedient employee perhaps went beyond what Bauer J observed in the Maine Lighting case (although Bauer J did refer to "foolish disregard for personal safety" in the context of human error) but reflected what Dixon CJ in Dunlop Rubber Australia Ltd v Buckley, a case concerning the duty of employer to securely fence dangerous machinery, had extracted from a judgment of Lord Justice Denning in Smithwick v National Coal Board (1950) 2 KB 335, also in relation to the fencing of dangerous machinery. Denning LJ said: The occupier must realize that not everybody is careful: many are hasty, careless or inadvertent; some are unreasonable, or even disobedient. It may be unlikely that they will act in such a way, but it is not only the likely but also the unlikely accident against which the occupier must guard. He must guard against all conduct which he can reasonably foresee. The limit of his responsibility is only reached when the machinery is safe for all except the incalculable individual against whom no reasonable foresight can provide - the individual who does not merely do what is unlikely, but also what is unforeseeable, or, at least, not to be foreseen by any ordinary man" ((1950) 2 KB 335 at 350).

154 I would take the reference to reasonable foreseeability in the Arbor Products judgment not to be intended to introduce an element in the offence but to be referring to the defences available under s 53. If it were otherwise, there is no basis for concluding Powercoal could have reasonably foreseen the skimming of stook X. The mining plans were predicated on the leaving of a stook and, while the very existence of the terms "skimming" or "robbing" the stook might suggest the existence of some practice, the evidence did not seek to (and did not) establish any such practice, and thus foreseeability.

155 There was no suggestion in the case that Powercoal was seeking through its employees to avoid a statutorily-regulated mining scheme which emphasises safety and the duties of mining officials and mining crew members to comply with support rules and work safely. The risk of injury involved in this case arose from an event the cause of which was not able to be ascertained from the evidence.

156 I conclude the prosecution has not established a relevant failure by Powercoal in relation to the recording and notification charge.

157 It follows that the charge against Mr Foster must be found to have not been established.

158 I would dismiss the charges in each matter. I will hear the parties on costs if it be necessary.

DISCLAIMER - Every effort has been made to comply with suppression orders or statutory provisions prohibiting publication that may apply to this judgment or decision. The onus remains on any person using material in the judgment or decision to ensure that the intended use of that material does not breach any such order or provision. Further enquiries may be directed to the Registry of the Court or Tribunal in which it was generated.