Incident directory

1996 - Leos Supermarket




  • Building Fires




Date of event

4th February 1996                 

Time of event


Name of premises

Leos Supermarket.


Broad Street, Staple Hill, Bristol.

Service area

Avon Fire Brigade (AFB) now Avon FRS (AFRS).

Nature of incident


Property type

Shop/warehouse approximately 60 m x 40 m in an L shape consisting of part single and part two storeys.

Premises use


Construction type and materials

Some steel frame, brick walls with a corrugated asbestos roof over a concrete base. First flooring materials of timber.


Customers and staff.

Fire source and location of fire

Deliberate ignition involving polystyrene packaging materials in the ground floor meat preparation area.


Brief Synopsis

Avon Fire Brigade (now Avon Fire and Rescue Service (AFRS)) were called to a fire at 12:46 on the 4th February 1996 at Leo’s Supermarket, Staple Hill, Bristol. The initial 2 pump pre-determined attendance arrived at 12:51 and 12:52 and the initial officer in charge (OIC) a Sub Officer (SubO 1) almost immediately made pumps 4 (AFB Investigation Team, 1996).

The first of 2 entry control points (ECP) was established with ECP 1 set up by the side entrance used as a fire exit, where 2 jets and then 2 teams of 2 BA were got to work. A Station Officer (StnO 1) arrived at approximately 12:57 and took charge at approximately 12:59. An Assistant Divisional Officer (ADO 1) arrived at approximately 13:01 as was in charge a few minutes later (AFB Investigation Team, 1996).

ECP 2 was set up by the main entrance where 2 further teams of 2 BA were committed at essentially the same time, initially intended to work together as a team of 4. Team 1 on ECP 2 (including Ff Lombard) was committed with a communication radio and a guideline to lay, where team 2 on ECP 2 was committed with a hose reel jet. The 4 BA wearers entered the premises at approximately 13:06. The 2 BA teams separated soon after entry to the fire (AFB Investigation Team, 1996).

The fire developed and team 2 who were closer to the fire, had to withdraw due to the deteriorating conditions. Team 1 who were initially heading away from the fire had also decided to withdraw due to an evacuation signal at approximately 13:11. Their exit route took them back towards the fire, and they were caught in some sort of rapid fire development discussed as a backdraft, flashover and as suggested within the service investigation report, an area of ‘local deepening’. Ff Fleur Lombard was killed (AFB Investigation Team, 1996).

Further assistance messages were received by control at 13:11 (make pumps 6 and hydraulic platform required) with a breathing apparatus (BA) emergency message declared at approximately 13:15 and at 13:43 a make pumps 10 assistance message. A stop message was sent at approximately 16:43 and the incident was closed at approximately 18:04 (AFB Investigation Team, 1996).

Photo 1

Images courtesy of Avon Fire and Rescue Service (AFRS).

Photo 2

Ground floor plan. Images courtesy of Avon Fire and Rescue Service (AFRS).

Photo 3

 First floor and roof. Images courtesy of Avon Fire and Rescue Service (AFRS).

Photo 4

BA team routes. Images courtesy of Avon Fire and Rescue Service (AFRS).

Photo 5

3D plan of ground floor. Images courtesy of Avon Fire and Rescue Service (AFRS).

Photo 6

Images courtesy of Avon Fire and Rescue Service (AFRS).

Photo 7

External image. Images courtesy of Avon Fire and Rescue Service (AFRS).

Photo 8

External image. Images courtesy of Avon Fire and Rescue Service (AFRS).

Photo 9

External image. Images courtesy of Avon Fire and Rescue Service (AFRS).

Photo 10

Images courtesy of Avon Fire and Rescue Service (AFRS).

Photo 11

Images courtesy of Avon Fire and Rescue Service (AFRS).

Main findings, key lessons & areas for learning

Avon Fire & Rescue Service summary of main conclusions & recommendations

Taken from: Avon Fire Brigade Investigation Team. (1996). Report of the Brigade Investigation into the Death of Firefighter Fleur Lombard. 


12  Conclusions and Recommendations

12.1  General

12.1.1  The conclusions have been primarily motivated by the Terms of Reference for the Investigation which were detailed in Paragraph 2.6.1. The Report hitherto has recorded all material facts revealed by the Investigation together with the relevant policies, practices, procedures and standards which did or should have applied.  In addition to establishing the circumstances surrounding the death of Firefighter Fleur Lombard, all other matters referred to in this Report have been the subject of investigation in an endeavour to identify contributory factors affecting the behavior of the fire and to identify any Brigade practices or procedures which may need to be reviewed as a result. Those findings and any recommendations arising follow in this section. Any conclusions or recommendations having implications outside the Brigade enable the Chief Fire Officer to consider whether he would wish to commend those matters to the relevant authorities.

12.1.2  For ease of reference, the conclusions have been addressed in five separate groups comprising Personnel, Breathing Apparatus, Command and Control, Training, and Fire Safety.

12.2  Personnel

12.2.1  Firefighter Fleur Lombard died from shock and extensive burns believed to have been caused by 'local deepening' effects of a massive flashover producing an estimated heat output of 8 to 10 MW at temperatures in excess of 1000°C which were sustained for a period of up to three minutes. Those conditions exceeded the design specifications of the personal protection uniform and breathing apparatus by up to four times thereby rendering the protection provided ineffective.

The operation in which Fleur was engaged with her colleague, Firefighter 9 at the time of her death was considered to be a legitimate and justifiable one. The command decision taken to commit the  breathing apparatus wearers into the building was based on sound and reasonable grounds considering that the initial  fire appeared localised in a distant part of the supermarket and was being controlled by jets in that location. The objective of the decision was to penetrate a smoke-logged part of the shop to lay a guideline and contain the fire from spreading into the unaffected area whilst searching for missing persons.

The development of conditions symptomatic of a flashover would not normally be considered likely when a fire penetrates the roof as it had at Leo's when the first crews arrived.  The flames and rising heat in such circumstances would usually indicate that the flammable gases present were being drawn into the fire, ignited and discharged through the roof. As has now been proven with such tragic consequences, these were not 'normal' conditions and the flashover that did occur with such devastating effect could not have been foreseen when the breathing apparatus crews were first committed. When potential flashover conditions became apparent; (the rising temperature and increasing noise), the evacuation of crews had already been initiated and the appropriate actions were taken.

The tragic and unfortunate death of Fleur Lombard was the ultimate sacrifice of someone who was carrying out instructions and doing her job to the best of her ability. She was trained to the appropriate standards to fulfil that role and even excelled in her competence. Both Fleur and Ff 9 were relatively relaxed about the tasks that lay ahead when they first entered the building and even exchanged banter between themselves as they progressed. Neither had been told anything different after they were approached by the Police Constable before entry, who told them he was not sure if everyone was out of the building. They were acting with the knowledge that customers might still have been in the supermarket which always stimulates that additional motivation and impetus to stretch the limits of personal endurance that only a professional firefighter would know. Whilst becoming aware of the deteriorating conditions inside when Ff 9 said to Fleur that they would not be able to stay in much longer because of the heat; they were still both determined to complete their task and tie-off the guide line at the end of the fateful aisle before retreating when, coincidentally, it is thought that the evacuation signal was received by Fleur from the BAECO. In any normal circumstances, both Fleur and Ff 9 would have withdrawn from the building safely to 'notch up' yet another smokey breathing apparatus 'job'. Inauspiciously, these were not normal circumstances but an inexplicable catastrophe which consumed in its wake, a very innocent victim. Both Fleur Lombard and Ff 9 displayed gallantry equal to the finest acts of bravery recorded and acted in the very highest traditions of the fire service whose primary function is to save life. Ff 9, when he became aware of his orientation after the flashover, showed a concern for his colleague which was instinctive and immediate and which prompted an instant reaction to go to her aid with absolutely no concern for his personal safety.

We RECOMMEND that the undaunted acts of valiance displayed by Fleur Lombard and Ff 9 be considered for a suitable award which appropriately recognises their actions and which could, in the case of Fleur Lombard, be awarded posthumously.

12.2.2  The courageous and gallant actions of Fire­ fighter 7 in immediately going to the assistance of Ff 9 in effecting the rescue of Fleur Lombard should also be recognised in an appropriate  manner and

We RECOMMEND that a suitable award that reflects such courageous and gallant actions should be considered for Firefighter 7.

12.2.3  For displaying initiative and fortitude as the breathing apparatus emergency was  unfolding, by initiating evacuation, by assuming command of Entry Control Point 2 when the emergency became apparent and by re-motivating firefighting operations amongst devastated firefighters after recovery of the deceased,

We RECOMMEND that Leading Firefighter L/Ff 1 be considered for a suitable commendation for his actions.

12.2.4  In singling out the above named individuals we recognise that the operations were a team effort in which several other personnel played a key role. We do not underestimate those efforts in extremely harrowing and distressing circumstances and we acknowledge the valorous actions of all personnel involved with the incident, particularly those directly involved with the breathing apparatus emergency.

12.2.5  We reported on our enquiries concerning the general application of breathing apparatus procedures in Paragraph 10.6.1 in which we recognised and acknowledged that operational decisions are taken on the basis of calculated, professional judgements made in the prevailing circumstances at the time. We also recorded how these cannot always be reconstructed after the event in a way which shows the imperatives perceived by those involved. In the minds of the Officers involved at the time, there will always be doubt about the wisdom of their actions and in particular, the veracity of the decision to commit breathing apparatus wearers into the building at all. We wish to record that we have found no evidence which has caused us to challenge or question the command decisions taken to commit the breathing apparatus wearers although we have comments of a more general nature concerning breathing apparatus procedures and command and control in the appropriate sections which follow.

12.2.6  Through the course of our enquires in conducting this investigation, it has been necessary to interview at length, the individuals in attendance at the incident at the critical time.

The incident had a profound effect on all members of the Brigade and many of those directly involved, some of whom were also immediate friends and colleagues of Fleur, were clearly suffering the effects of the trauma in an extremely distressing way. We pay tribute to the courage of those who assisted with our enquiries for their forthrightness, honesty and openness which enabled us to reconstruct the events with accuracy and confidence.

12.3     Breathing Apparatus Procedures

12.3.1  Of the two teams of breathing apparatus wearers who entered the premises from Entry Control Point 2, only Firefighter Lombard was equipped with 'Sonic' radio communications equipment. Had Teams 1 and 2 been able to communicate with each other they may not have become separated as they did and therefore the hosereel being carried by Team 2 would have been available for the protection of all wearers whilst they were inside. There is an inadequate provision of radio communications equipment for breathing   apparatus. Each pumping appliance within the Brigade carries only one 'Sonic' equipped breathing apparatus set amongst  the  four breathing apparatus sets that are issued to each appliance. Due to a current shortage, two appliances within the Brigade have no communications equipment for breathing apparatus issued at the time of preparing this Report. As a breathing apparatus team normally consists of two people it is considered that two communication sets should be provided for each pumping appliance.

We RECOMMEND that the provision of at least one set of radio communications equipment for every breathing apparatus team be considered a priority for the Brigade.

12.3.2  In Paragraphs 9.3.1 and 9.3.2 we discussed the term 'Breathing Apparatus Emergency' and highlighted the confusion which was apparent in the comprehension of the terminology and the ambiguity of the guidance available. No clear definitions appear to exist which explain what precisely is meant by the terms 'breathing apparatus emergency' 'wearer in distress', 'distress to wearer; or 'evacuation procedure'. The fact that we elicited a range of very different explanations of those terms from qualified breathing apparatus wearers and indeed, from qualified instructors would seem to suggest that greater clarity is required.

Exactly what constitutes a 'breathing apparatus emergency’ does not appear to be any more easily understood either. The term appears in Technical Bulletin 1/89 to describe a message which should be initiated by a Stage I Entry Control Officer in the event of 'emergency' conditions arising (Section 23(1)(i)) as a code for Brigade Control to mobilise reinforcements.  By way of comparison to those duties, the duty of a Stage II Entry Control Officer in the event of 'emergency' conditions arising is to "initiate emergency measures". It is assumed that the difference between the two pieces of guidance reflects the presence of an 'emergency' team under Stage II Entry Control procedure which may not be standing by under Stage I where (in Avon Brigade) less than six wearers are deployed.

The term 'breathing apparatus emergency' is also common parlance to describe the conditions which are declared in the event of a 'wearer in distress' or 'distress to wearer'. In the absence of a clear definition of these terms however, a 'breathing apparatus emergency' appears to be automatically declared following the 'distressing' condition giving rise to it. The additional activities which should be triggered by a 'breathing apparatus emergency' declaration are different under Stage I and Stage II which adds to the confusion and misunderstanding of the terminology. The actual use of the word 'emergency' throughout the  guidance  is ambiguous in some contexts and appears to apply to all abnormal breathing apparatus conditions whether a 'real' emergency or not. Whilst recognising that this was probably intentional as a failsafe precaution when the guidance was prepared, an actual 'breathing apparatus emergency' procedure is not always justified or necessary. In the absence of any distinction between an 'urgent' and 'non-urgent' malfunction or distressing condition, the full blown emergency procedures apply in all cases which may dilute the urgency and importance which should be associated with a life-threatening  emergency condition.

The provision of additional breathing apparatus sets on fire appliances in recent years has added greatly to the ability to implement emergency procedures in Stage I for first attendance appliances which has not been recognised. The requirements for reporting 'notifiable Dangerous Occurrences' under RIDDOR which amended  Technical Bulletin 1/89 in 1991 further confuses the perceived understanding of a 'breathing apparatus emergency' and a 'notifiable Dangerous Occurrence'. Whilst those conditions may frequently conform to a similar definition, it does not apply in all cases. The guidance available in this regard appears to be driven mainly by the reporting requirements and the needs of a subsequent investigation into the cause of the malfunction rather than the retrieval and recovery of the distressed breathing apparatus wearer. The implications of the Management of Health and Safety at Work Regulations 1992 may well influence a plan of action for the recovery of a distressed wearer as part of the risk assessment process.

In the light of the apparent confusion about the terminology used in respect of breathing apparatus emergencies,

We RECOMMEND that the available guidance be reviewed to consider what, if any, revisions may be appropriate to rectify the following perceived deficiencies:

(a)  the absence of definition or clarity of the terms "Breathing Apparatus Emergency", "wearer in distress" and "distress to wearer" (or any alternatives to those terms);

(b)  the absence of any precise actions which should be followed or devised in the event of those conditions arising and in what circumstances they are implemented';

(c)  the inconsistency of guidance under Stage I and Stage II breathing apparatus control in 'emergency' conditions;

(d)  a clear distinction between measures necessary to retrieve a situation or recover a distressed wearer and the needs for reporting or subsequent investigation, and;

(e)  the absence of any recognition of a difference between "urgent" and "non-urgent" emergency conditions which may require different levels of action.

12.3.3  We discussed in Paragraph 8.7.1 and 8.7.2 how the phenomenon known as 'local deepening' would have evolved   in   the   flashover   condition   that   killed Fleur Lombard. We also summarised, on the advice of the Fire Research Station that the deepening layer of flame temperature gases at ceiling height would curl down a vertical obstruction towards floor level. It is the assumed behavior of those thermal flows that we believe caused such devastating injuries to Fleur who was adjacent to the vertical obstruction. Ff 9, who was slightly further away, escaped the worst effects of the thermal flows but was thrown to the floor by the blast. If the hypothesis is correct, there could be implications on the advice given to breathing apparatus wearers to maintain contact with a side wall when searching in smoke.

We RECOMMEND that the phenomenon 'local deepening' be further investigated to determine any implications for firefighters, particularly when searching or plotting a course in a smoke filled building.

12.3.4  We recommended in Paragraph 12.3.1 above, the extended provision of radio     communications equipment to ensure that a set is available for every team. With  the  exception  of  a reference  in  a  paper  attached  to 'Dear Chief Officer letter 4/1988, no  guidance  has  been found  which  sets  out  a  critique in respect of breathing apparatus  communications  procedures. We have already commented  on  the  fact  that  the  two  teams  of  breathing apparatus  wearers  who  entered  through  Entry Control Point 2 may not have become separated had they both been equipped  with communications  between  them. We also believe that intelligence about the deteriorating conditions inside the building  prior to the fatal flashover could have been  exchanged to  better  advantage both to  and from the teams as information became available. Whilst it would be speculation   to   suggest that the availability of such intelligence in the Leos incident could have led to an earlier withdrawal, the changing conditions would undoubtedly have been better known to all concerned in respect of very basic risk assessment.

We RECOMMEND that consideration be given to the development of a communication protocol between each team and their breathing apparatus Control to monitor progress of the team, conditions inside, intelligence about risks and hazards, other firefighting activities in progress and about information which the team inside the risk should know about.

12.3.5  We recognise in making the above recommendation, that such a procedure would place additional burdens on the existing responsibilities of BAECO's, particularly in Stage I breathing apparatus Control. The existing duties of the BAECO are already considered onerous enough for what could be a very junior individual especially if that responsibility should extend to initiating emergency procedures or deciding on an evacuation due to changing conditions. Any additions to existing responsibilities could well detract from the effective discharge of those duties and as a result, undermine the primary purpose of managing and controlling their breathing apparatus Control Board.

We RECOMMEND that an enhancement of existing procedures should be considered to provide for a dedicated Officer-in-Charge of a breathing apparatus control point, as and when resources permit, who should be a junior officer rank and whose principal duties would include:

(a)  maintaining radio communication with teams working from that entry point;

(b)  logging intelligence gathered as a result of that communication and passing it on as appropriate;

(c)  taking responsibility from the BAECO for command decisions which may arise;

(d)  maintaining a risk assessment over­ view to inform or initiate actions arising out of information received.

The appropriate time during the progress of an incident at which resources would permit this escalation may well coincide with the existing guidance about the provision of an emergency team - i.e. as and when resources permit or when the number of wearers at an entry point exceeds ten (six in Avon Fire  Brigade).

12.3.6  We reported that significant setbacks to the investigation process were the failure to recover and impound the breathing apparatus set worn by Fleur Lombard, the delay in impounding the breathing apparatus control board causing it to be eventually impounded in an incomplete state, the disposal by the hospital authorities of Fleur Lombard's personal kit and the diversion to other duties of the breathing apparatus emergency Supervisory Officer. These were a catalogue of misfortunes which have not been impossible to overcome during the investigation but nonetheless should not have arisen if strict procedures had been followed.

We RECOMMEND that a 'System of Work' be devised which sets out the exact procedures to be  followed in addition  to  and  separate  from any procedures arising from the Recommendations in Paragraph 12.3.2 above in the event of a 'reportable Dangerous Occurrence'. The System of Work should include detailed information about impounding equipment, preserving and photographing evidence, individual responsibilities, and supervision of the process, advice for incident commanders, the supervision of casualties and the reporting requirements. The 'System of Work' should be readily available on the fireground for the reference of the Supervisory Officer.

12.3.7  We referred to discrepancies in the existing advice given in respect of the above in the Brigade's O & T Note A24 and the Appliance and Equipment Manual, Section B7.

We RECOMMEND that the O & T Note on breathing apparatus procedures be reviewed to address the discrepancies with the A & E Manual, and at the same time, to ensure that the O & T Note accurately reflects in sufficient detail, exactly what the Brigade's policies are where, either they exceed those outlined in Technical Bulletin 1/89, or where the guidance contained in the Technical Bulletin requires elaboration or tailoring to suit the needs of Avon Fire Brigade. The Recommendations in Paragraph 12.3.2 above could be examples of the need for local clarification should this not be considered appropriate on a national basis.

12.3.8  Although we make no criticism of Fleur Lombard or Ff 9 in taking the actions they did, we could not overlook the part experience played in this tragic sequence of events and consider whether prevailing conditions could have influenced the outcome. Fleur Lombard had two years and three months service and Ff 9 about four and a half years. At the time we believe Fleur received the message to evacuate from the BAECO and conveyed that to Ff 9 by shouting "Evacuate, evacuate", Ff 9 had already recognised the worsening conditions and told Fleur that "they wouldn't be able to stay in much longer". The decision to evacuate would probably have been taken by them anyway had the message not been received.

It is debatable whether a firefighter with much longer service than these two would have experienced or known how to recognise the symptoms of a serious deflagration given the speed of events in this case. We cannot therefore conclude with any conviction that greater experience would have influenced the outcome.

However, in the light of our recommendation for the development of a breathing apparatus communications protocol (Paragraph 12.3.4) it would appear sensible to have the more experienced member of the team fitted with the communications equipment.

We RECOMMEND that consideration be given to revising the guidance in this respect to provide for the senior rank or most experienced member of the team to assume the role as 'team leader'.

12.4  Command and Control

12.4.1  We have highlighted the confusion that arose when Teams 1 and 2 made their initial entry to the building from Entry Control Point 2. Team 1 (Lombard/Ff 9) were unsure whether they were expected to take the hosereel as well as laying the guideline. Firefighter 9 had to return to the BAECO to check that they had been detailed to lay the guideline to the left.  Firefighter Lombard when entering, began to proceed to the right before being corrected by her colleague. Team 2 (Ff 7/Ff 5) were not aware that the two teams should have been working together or in what direction they were supposed to proceed. Neither of the teams were aware that other breathing apparatus wearers were working in another part of the shop and using jets, and those breathing apparatus wearers were not told that further teams were entering the building from a different entry point. The BAECO only became aware of the breathing apparatus teams' tasks by coincidence and was not formally advised of the command strategy or briefing given.

All of this confusion could have been avoided through better briefing which in this instance proved woefully inadequate. That the fatal outcome of the incident did not appear to have been influenced by the poor briefing was fortuitous and very little comfort can be derived from that. Briefing of breathing apparatus wearers prior to entering the risk area is a basic but essential command and control function which was not afforded a sufficiently high level of priority. A communications protocol and a dedicated Officer-in-Charge of the entry control point as recommended in Paragraphs 12.3.4 and 12.3.5 respectively could well have overcome many of the briefing deficiencies for the breathing apparatus teams.

We RECOMMEND that the importance of good briefing and information exchange in both directions be given a higher profile in routine and refresher breathing apparatus training to ensure that breathing apparatus wearers in particular are made aware of all relevant information to provide them with better control over their own risk management decisions and others involved in the control and management of breathing apparatus wearers are better acquainted with fireground events and command strategy.

12.4.2  During the first ten minutes into the incident, the role of Officer-in-Charge passed on no less than four times due to 'home ground' rules or seniority of rank. This was a clear impediment to the development of a cohesive command and control strategy by either officer until well into the incident. The early stages of any fire incident are crucial and command decisions taken at that time have a major impact on evolving conditions. A possible alternative could be to continue fireground command under the responsibility of the senior officer on arrival up to a pre-determined point whilst more senior officers attending subsequently would assume the role of incident commander.

We RECOMMEND that the question of incident command in these circumstances be reconsidered with a view to allowing a cohesive command and control strategy to be developed in the early stages of an incident.

12.4.3  In our endeavours to reconstruct the incident in Section 5 of the Report, we became aware of the great distances travelled by officers around the fireground either in pursuit of another officer or to reconnoitre progress in other parts. Overall control of the incident can be diminished by the need to be in so many different places at the same time and the assimilation of information suffers as a result. In the Leos incident, we reported how the Officer­ in-Charge first became aware of the fatality at the incident when he was asked by an ambulance paramedic if he wished to view the body. As a consequence of these factors, we are of the view that the majority of movements and intelligence needs could be eradicated by effective communication facilities. Currently, the provision of hand-held personal radios available on the fireground is restricted to two per pumping appliance with five on each of the Control Unit and the Chemical Incident Unit and two on each Rescue Tender. At an incident such as Leo's, where only pumping appliances would be in attendance in the early stages, the majority of available radios would be held by the appliance Officers-in-Charge and their contact point - usually their driver. This leaves an inadequate provision for other crew members or for senior officers arriving at the incident subsequently who therefore have to resort to making personal contact with their subordinates or commandeering a radio from an existing holder which depletes their communications ability.

We RECOMMEND that consideration be given to the provision of additional personal radios as soon as possible to increase the number available on pumping appliances and to ensure that all officers arriving at an incident have personal access to one.

12.4.4  We reported on the fact that a police constable in attendance at the Leo's incident had informed Firefighters Lombard and 9 before entry that he was not sure if everyone in the supermarket had been accounted for.  We also reported that the police constable had been directed towards the Station Officer-in-Charge and told to inform him of his concerns.  The police constable in the event, was unable to make immediate contact with the Station Officer due to his preoccupation  with fireground command and it was  not for several minutes  that  the police  constable  was able to express his concerns to the Station Officer.  Either because  the  concerns  were  expressed  differently  to  the Station Officer than they had  been  to Team  1, or because they were interpreted  differently by the Station Officer is not  clear. Whatever the position, the Station Officer appeared to understand that the police constable's concerns applied in respect of the Labour Club rather than the supermarket. When the Station Officer conferred about this with the Sub Officer, the Sub Officer confirmed that the Labour Club had been evacuated. This appears to be a clear example of misunderstanding which causes concern. The police constable's concerns that customers might still be in the supermarket as expressed to Fleur Lombard and Ff 9 were in fact, ill founded and based on his conversation with the duty manager of the supermarket who had been unable to confirm that everyone was out. The police constable, quite rightly, considered this to be information which the Brigade ought to know about and he, quite rightly, was directed to the Officer-in-Charge. The fact that this information had not been countermanded, contradicted or retracted with Team 1 was discussed in Paragraph 12.2.1 where appropriate recommendations were made. Our main concern arises in respect of a different scenario if the information about persons inside the building was accurate. The misunderstandings in these circumstances could have had much more serious implications and therefore underlines the importance of intelligence gathering at any incident. We make no specific recommendation on this issue other than the need for greater emphasis to be placed on briefing during training which has already been referred to but urge that relevant lessons be learned from these events which also serve to reinforce the need for improved fireground communications.

12.5  Training

12.5.1  The standards of recruit training to which all new entrants to the fire service are subjected was found to be in conformity with current guidance - Fire Service Circular 7/1974. The basic breathing apparatus training also conformed to the current standards laid down in Fire Service Circular 8/ 1981 and it has been noted that Fleur Lombard excelled in both areas and achieved the top marks on her recruits course to win the coveted 'Silver Axe Award'. Our investigation has not addressed the adequacy of the training in the present day environment, as the ability and competence of the breathing apparatus wearers in the performance of their tasks has not been an issue. We do however note that the recommended Wholetime Recruit Training Syllabus is 26 years old and the recommended breathing apparatus Training Syllabus is 15 years old. Whether those syllabi still reflect the training needs of todays firefighters and whether the latest knowledge and technology is adequately satisfied by the training are not questions that this investigation should have answered. We do however feel that the absence of real fire training in the breathing apparatus syllabus - i.e. using breathing apparatus in realistic conditions of heat and smoke rather than simulated conditions using cosmetically generated smoke, is an issue which may need to be reconsidered in the light of this incident. Although we do not cast any aspersions on the credibility of Fleur Lombard and Ff 9's experience of breathing· apparatus wearing in real fire conditions, exposure to such conditions in a training environment has to have more positive than negative effects on the completeness and preparedness of the individuals undergoing that training. Ff 9 had the benefit of the Realistic Fire Training course undertaken at the Fire Service College whilst Fleur Lombard did not.

To what extent that additional training experience may have affected the outcome at this incident for Fleur Lombard is a matter of speculation. Under the present training syllabus it would be theoretically possible for a newly qualified breathing apparatus wearer to emerge from training school, to be appointed to a watch as a trained member of a crew and attend as their first 'breathing apparatus job', another Leo's fire. It has been a source of some relief to report that newly qualified firefighter recruits in Avon now attend the Realistic Fire Training course at the Fire Service College during the probationary period although financial resources prohibited this during Fleur Lombard's training. We believe this additional exposure to realistic conditions in a controlled training environment to be an essential element of training which should continue. There are however, no facilities within the Brigade for real fire training which prohibits any continuation training either for routine or refresher training needs or for qualified firefighters who have never had such training. There are also inadequate facilities within the Brigade for training in simulated heat and smoke which places heavy demands on the one facility that does exist and creates an inefficient and logistical nightmare to arrange annual breathing apparatus refresher courses for the whole Brigade at that one location.

We RECOMMEND that priority be given to the provision of better facilities to improve the standards of breathing apparatus training in both simulated and real fire conditions as a matter of urgency.

12.5.2  We referred to a number of instances where prescribed procedures were not strictly observed in several areas of breathing apparatus wearing and control procedures. Our recommendation in Paragraph 12.4.1 to raise the profile of good briefing during training was one of those areas which we felt required specific attention for command and control reasons and to ensure that the breathing apparatus wearers themselves have the best quality information available to have greater control over their own safety assessments and decisions. To be wholly effective however, the wearers themselves have to fulfil their obligations by observing the established procedures developed for their own safety and that of their colleagues. The lack of rigor towards breathing apparatus procedures that is occasionally witnessed can be difficult to comprehend to the uninitiated. If viewed in the context of the screaming mother approaching breathing apparatus wearers on arrival at an incident that her child is trapped inside the fire, any mitigation of procedures becomes more understandable. The non-observance of safety procedures in such circumstances is not through any blatant disregard of the rules but usually because of the instinctive urgency to do their job, often in the face of inadequate resources to fulfil all of the needs at the time. The pressures on those concerned at the Leo's incident has not been underestimated and we believe that the oversights in respect of breathing apparatus procedures were more unfortunate than deliberate. We have no grounds to believe that any procedural inadequacies affected the outcome of the incident or contributed in any way to_ the death of Fleur Lombard. In attempting to understand the circumstances under which the omissions arose, it would nonetheless be remiss of this Report not to draw attention to these matters.

We RECOMMEND that appropriate steps are taken to:

(a)  reiterate through training, the need to observe and practise the established policies and procedures on all occasions that breathing apparatus is worn;

(b)  to ensure at all times, that adequate resources are provided by the Brigade to facilitate observance of the procedures.

12.5.3  We remarked in the Report on the duties of the BAECO and we have made a recommendation (Paragraph 12.3.5) to enhance the existing breathing apparatus procedures to provide for a dedicated Officer-in-Charge of a breathing apparatus control point. Officers at all levels would normally only exercise a breathing apparatus controlling function as part of their command and control role fairly infrequently since the BAECO would more often than not be a member of an appliance crew rather than an officer. The overall management of breathing apparatus control by officers is a function which normally applies as breathing apparatus operations escalate to Stage II or Main .Control. The infrequency of such incidents provides insufficient opportunities for the management role to be practised and the existing limitations on resources make it extremely difficult to include such training within the refresher programme. For a large scale breathing apparatus operation to run efficiently we believe it to be vitally important that training scenarios should include the relevant level of officering at all scales of the incident to create the necessary realism and to provide ongoing refresher training opportunities for the officers themselves.

We RECOMMEND that as soon as facilities and resources allow, training opportunities be provided to allow officers at the appropriate levels to practise the management of breathing apparatus control on a regular basis in line with the realistic expectations of their potential role in the command and control of breathing apparatus  operations.

12.5.4  In Paragraph 10.7.3 of the Report, we outlined in some detail the Probationary Training Scheme to which Fleur Lombard had · been conditioned during her probationary period. We reported on how the scheme had been devised as a continual assessment of objective performance across a range of competencies which were weighted to give balance and priority to different key result areas. Whilst not directly associated with the incident at Leo's, we did feel that it was encompassed within the Terms of Reference to note that revisions have subsequently been made to the Probationary Training Scheme since Fleur Lombard completed her recruit training that diminish the measurement of competence rather than enhance it. We feel that a comprehensive, competence based evaluation of training received is the most appropriate method of validating competence levels of students particularly in the field of recruit and probationary training.

We RECOMMEND that the former concept of competence based validation of recruit and probationary training be reinstated as a matter of urgency.

12.5.5  We commented in Paragraph 10.7.8 (Training in the Phenomenon of Flashovers) how the latest available information on the subject had been published in 1995 as a supplement to Books 1 and 11 of the Manuals of Firemanship. Although we were not able to establish whether publication of the supplement influenced routine training programmes, we commented how no formal advice had been offered by the Brigade in respect of any actions which need to be followed on receipt of the new publication. This question opened a more general line of enquiry which extended to the way in which new policies, procedures and specific information is communicated to operational personnel. We were of the opinion that in the general sense, information is communicated fairly effectively through the mediums of existing Brigade promulgations and ongoing refinements of those procedures are streamlining the process even further. We did however, have reservations about the way in which the supplement had been issued to stations with no guidance about how its contents might influence training. Although the publication of supplements in that form is a new concept adopted by the Home Office to keep brigades updated on new developments, the way the new information is handled was equally important and we have been pleased to note that a new approach has been adopted by the Brigade as a direct result of this finding.

We RECOMMEND that incoming  information having operational implications becomes the subject of scrutiny to assess any impact on practices, policies or procedures about which specific advice should be offered and in particular, about any training requirements arising as a result.

12.5.6  In arriving at the foregoing conclusions in respect of training, we have highlighted specific aspects about which we have made appropriate recommendations. Insofar as training generally is concerned, we are aware of the difficulties the Brigade faces in fulfilling all internal training needs which involve taking personnel 'off-the-run' to undergo the training due to ridership factors. The Brigade's wholetime establishment provides only sufficient personnel to meet minimum crewing levels on appliances. In the absence of an excess allowance for sickness, training or other unplanned absences, the ability to remove people from operational duties for training purposes is severely curtailed, particularly when the establishment level falls below the minimum (as it frequently does in between recruit intakes) or when an excessively high number of personnel are absent on long-term sickness. This will continue to present a problem for the delivery of effective, structured training to meet future ongoing needs and to make significant improvements to the existing training provision. Many of the shortcomings in procedure highlighted in this Report relate to breathing apparatus procedures which have a training implication. Whilst the Brigade's establishment remains at its present levels, it is difficult to envisage how major improvements could be achieved to the absolute minimum level of breathing apparatus refresher training provided at the present without affecting the ridership of operational fire appliances.

We RECOMMEND that further consideration be given to the Brigade's wholetime establishment level to examine ways of achieving a training allowance as soon as circumstances permit.

12.6  Fire Safety

12.6.1  We quoted in Paragraph 3.6.4 the requirements incumbent upon an occupier of a premise in respect of which a Fire Certificate was in force under Section 29 of the OSRP Act. We also drew attention to the following matters which arose as a consequence of our investigation at Leo's.

(a)  The manual fire alarm system installed in the premises was not operated when the fire was discovered.

(b)  The duty manager's spontaneous attempt to use the shop tannoy system to evacuate the store was foiled (Paragraph 5.2.4).

(c)  There was no ordered or disciplined approach to the evacuation of customers from the shop by staff after the fire was discovered (Paragraph 5.2.3).

(d)  The contracted security guard had not been familiarised with the building or the fire procedure when he reported for duty at the premises for the first time (Paragraph 5.1.2).

(e)  A member of staff was unable to tell the fire brigade the address of the premises during the initial 999 call (Paragraph 5.2.4).

(f)  The duty manager had to break out of the staff room window despite the fact that an exit door from that area showed a means of egress to open air (Paragraph 5.3).

Taken as a whole, these matters did not engender high levels of confidence in the fire safety management culture prevailing at the store when the fire occurred. We were particularly concerned to establish whether the misinformation about the possibility of customers remaining in the

The store which was passed from the duty manager, to the police constable and finally to the fire officer could have been avoided. The absence of a cohesive evacuation procedure which attempted to chaperone and clear customers out of the shop led us to conclude that there had not been an ordered evacuation of the premises and the one checkout operator who did attempt to check the aisles for customers was acting on her own initiative. Had  an effective evacuation procedure been adopted to ensure that customers were directed towards exits and the aisles were systematically checked by staff, we feel it would  be unlikely that customers would have been reported as being unaccounted for inside. In such circumstances, Fleur Lombard and Ff 9 would not have heard the report about missing persons and we feel may not therefore have stretched their limits of endurance as they did.  To this extent we feel that the poor fire safety management procedures at the store could have influenced their actions and those of the officers in command of the    incident.

12.6.2  The plan of the premises which formed the First Schedule of the Fire Certificate and which remained current at the time of the fire showed that alterations had been made to the premises since the Certificate was issued which did not appear to have been the subject of consultation with the fire authority as required by Section 30(2) of the Offices, Shops and Railway Premises Act 1963. Alterations appeared to have been made in the 'Butchery' (referred to in the Report as the 'Meat Preparation Room') and in the area designated on the Certificate as the 'Cafeteria' (shown on the plans accompanying this Report as the Staff Room/Kitchen/Offices). Further alterations affecting the 'Stock Room' and 'Bond' were also made as part of substantial alterations to form a separate occupancy within the ground floor area which was occupied by a furniture retailer at the time of the fire. We have not been able to identify any consultation records in respect of either of those alterations even though consultation did take place concerning a proposal to create offices at first floor level in 1974. The main effect of the alterations on the ground floor undertaken apparently without consultation was to remove or alter the designation of some of the fire exit doors detailed on the Schedule to the Fire Certificate.

We RECOMMEND that this matter be made the subject of further inquiries to establish whether formal procedures were followed correctly and whether there was any breach of statutory requirements.

12.6.3  When drawings were submitted by the CRS to the fire authority in July 1972 showing proposed alterations to the premises, a full schedule of recommendations was issued by the fire brigade   in   response   on 1 August 1972. The recommendations related only to proposed alterations to the ground floor, the first floor not being proposed for use at that time. Various recommendations were made in respect of means of escape, fire appliances, fire alarm system and fire exit notices together with others under a 'General' heading.

Amongst them was:

"In view of the size of the building thought should be given to the installation of an automatic sprinkler system, which should be in strict compliance with the Fire Offices Committee Rules for automatic sprinkler installations together with the appropriate amendments".

The Schedule of Recommendations also intimated that the premises would become subject to the Offices, Shops and Railway Premises Act 1963 when the proposed alterations were completed.

No powers existed either then or now, under which the fire authority could make the installation of sprinklers compulsory in shop premises and the recommendation could only therefore be on a goodwill basis. However, under certain circumstances with retail complexes requiring smoke control management measures, the provision of sprinklers would be an essential, mandatory element of the (means for securing) means of escape package, but generally speaking, under current Building Regulations, sprinklers in shop premises are merely recommended in premises exceeding 7000m3 (BS 5588: Part 2: 1985). The total capacity of the whole ground floor at Leos, by way of comparison was approximately 4600m3 A sprinkler installation may be designed into a speculative building as a compensatory feature to satisfy a 'Functional Requirement' under Building Regulations when a different requirement is not achieved. Sprinklers are unlikely to be installed in smaller buildings similar to Leos on a voluntary basis therefore, unless it became a requirement or the incentives to do so were more persuasive. Incentives do currently exist through reduced insurance premiums and tax concessions on capital costs but these can to some extent be offset by the requirements of the water undertaking. The insurance industry would appear to hold the key to this dilemma and of course they frequently impose stiff requirements on occupiers as a condition of insuring the risk. The growing tendency, particularly for larger companies, to self-insure part of their risk to a pre­ determined loss value before re-insuring above set loss limits, tends to undermine the insurance incentives for installing sprinklers.

Ironically, the commentary to the guidance on sprinkler systems contained in BS 5588: Part 2, (Section 26.2.1) states:

"The success rate of sprinkler systems in containing fire outbreaks until the arrival of the fire brigade has been put as high as 95% in cases where the system was in operative order at the time of the fire. There appears to be virtually no weakness in these systems unless they are turned off for maintenance or building work, or through negligence".

The guidance available when the Certificate was issued to Leos in 1973 was very different from the present, but the incentives were in existence at that time. Whatever considerations were taken into account by the CRS at the time, they did not take up the recommendation of the fire authority to install sprinklers at the premises when the alterations were proposed in 1972.

We believe that had an automatic sprinkler system been installed at the time of the fire, early detection would have been achieved in that unoccupied meat preparation room where the fire started and the sprinklers would have controlled the spread and development of the fire before the fire brigade arrived and as a consequence, allowed the brigade a reasonable chance to contain and extinguish the fire and save the building. Such a theory could have prevented the loss of life as a result.

The enormous heat levels generated by the Leos fire were probably typical of the output that could be expected from a totally involved retail occupancy of this type. The estimated peak heat release of 8 to 10 Megawatts referred to in Paragraph 8.3.3., characterised the high fire loading which existed at the time of the fire and provides a sombre comparison to the 'standard' 5MW 'design fire size' used generally in calculations for the design of smoke (and sprinkler) control measures. A high proportion of the commodities stocked were of a combustible nature and the presence of 350 litres of motor oil in plastic containers together with the added combustibility presented by LPG charged aerosol containers and alcoholic spirits would certainly have increased the fire loading significantly.

Recent fires in other parts of the country in this type of occupancy have made their own contributions to the experiences being gained about high loss fires in large single compartment retail outlets.

Whilst clearly, these were different circumstances and whereas those more modern buildings allowed for total loss of the structure (due to relaxed compartmentation and fire resistance to elements of structure) but with state of the art means of escape and evacuation measures as compensation. The older, more traditionally constructed stores with similar fire loading have been subject to greater loss of life (Woolworths, Littlewoods etc) with less damage to the structure. In the Leo's circumstances we have the unfortunate combination of old design combined with lack of compartmentation and lack of fire resistance of elements of structure. Clearly in applying modern risk assessment techniques to the older retail premises with regard to life safety, some additional safety measures would be seen to be needed.

We believe there has never been greater justification of the need for sprinklers in supermarkets than is available on present evidence and experience. The Leos fire leads us to conclude that the 7000m3 benchmark recommended in BS 5588 for shops excludes a large number of small and medium occupancies of this type which have a high fire loading with a potential for large numbers of people present.

Many brigades have stepped back from even recommending sprinklers - in the knowledge that they are unlikely to be even considered for below 'cube' buildings and have gone for more economic 'traditional' preventative and detection measures.

Retrospective application of regulations has rarely been a feature of British law and on that basis, existing occupancies of a similar type would escape any requirement to install sprinklers if new regulations were introduced. For any significant improvements to be made in this situation, it appears to us that inducements to comply with a recommendation to install sprinklers must be far more attractive than at present. The question of sprinkler provision in retail outlets has been a controversial issue for some considerable time and it is understandable that the economic argument should prevail.  The imbalance that appears to exist between property protection and life protection in the assessment of risk by insurers seems a disincentive for the provision of passive fire protection measures for life protection.

We RECOMMEND that the criteria and incentives for the provision of automatic sprinkler installations in medium and small retail outlets having large undivided compartments be re-examined in the light of the fire at Leos and the conclusions drawn from it.

12.6.4  We outlined in Paragraph 10.7.9 the statutory obligations of the Brigade in respect of risk familiarisation under Section l(l)(d) of the Fire Services Act 1947. \Y/e also identified that whilst Leos would not qualify for risk information or familiarisation visits the Brigade's existing risk information system had not been maintained for several years. \Y/e stated that the Risk Information Card (Gen.3) had fallen into disrepute because of resource difficulties in maintaining accurate up-to-date information. A new 'Hazardous Premises' card system was devised to obviate the difficulties with the former risk information card but was never launched.

We also reported how the risk familiarisation visits are initiated and managed by the watch Officers-in-Charge. We feel that greater benefits could be derived from risk familiarisation, or '1(1)(d)' visits by clear guidance in a 'System of Work' which defines the  purpose,  aims, learning outcomes etc together with a definition of the target risk. It may be that this is or should be a matter which falls within the remit of the newly formed risk assessment sub-committee of the Safety Policy Management Team.

We RECOMMEND that a suitable system of risk information be resurrected or devised that includes a 'System of Work' in respect of risk familiarisation visits for operational crews and which embodies training needs as well as risk assessment needs.

12.6.5  The development of the Leos fire revealed a number of fire safety questions which may require further consideration or research to establish whether current compliance standards are adequate in the light of the experiences. We draw attention to the issues that arose in the course of the investigation in the following paragraphs which are described in more detail in the relevant parts of the Report. Whilst it was not considered to be within the scope of the investigation to delve too deeply into the technicalities of existing and past guidance on these matters, we single them out for special attention because we believe they were factors which contributed to the rapid development of the fire and therefore to the conditions that caused the flashover.

(a)  We stated in Paragraph 8.5.3. that we had no reason to believe that the fibreboard ceiling panels installed at Leos did not conform to the required surface spread of flame standard at the time of installation. The FRS however, suggested that the existence of the fibreboard ceiling with the continuous void above it would probably have been one of the crucial factors having a major contribution to the severity of the fire. The FRS also indicated that they have experience of ageing fibreboard degrading over time to Class 2 or 3 compared to the Class 1 that was originally specified. If the observations are correct, we would question the long term suitability of fibreboard as a false ceiling lining in public areas.

(b)  Whilst there would appear to have been no requirement for fire resisting separation between  the public accessed retail area and the staff  only warehouse area, this was also suggested as one of the crucial factors contributing to the severity of the fire. In view of the high fire loading within the building, the absence of automatic fire detection and the fact that the warehouse area was unoccupied, we feel that fire separation in this instance would have had a major effect on containing the fire.

(c)  As in (b) above, there appeared no requirement for vertical fire separation in the roof void to prevent the spread of heat and smoke throughout the entire premise. In view of the hybrid nature of the building and the consequent conglomeration of roofing lines and pitches we feel that the flow of hot gases in the roof void could have been affected by the various roof capacities which could have been influenced by compartmentation of the void. The extension of fire resisting separation between the retail and warehouse area as referred to in (b) above to true roof height would in itself we believe have had a significant effect on spread.

(d)  The lack of fire separation between the meat preparation area and the retail area has already been commented upon. The formation of the void between the two areas when the delicatessen was replaced by a display fixture is also considered an important issue because the void was used for storage. The decision to use it for the storage of crisps and savoury products however, was unfortunate due to the high calorific value of those commodities which was also a contributory factor in the development of the fire. We believe that better guidance may be appropriate in respect of the display and storage of crisps and savoury snacks in particular and indeed, other commodities of a highly combustible nature unless passive protection is provided.


FBU summary of main findings, key lessons & recommendations

Joint Avon Fire and Rescue Service/FBU report (please see above).

IFE Commentary & lessons if applicable

None produced at this time.

Known available source documents

Further information hoping to be identified and still to be located.

FRS Incident Report/s

Avon Fire Brigade Investigation Team. (1996) Report of the Brigade Investigation into the Death of Firefighter Fleur Lombard. [pdf]. Avon Fire Brigade.

FBU Incident Report/s

No information identified to date and/or still to be located.

Health & Safety Executive (HSE) Incident Report/s

No information identified to date and/or still to be located.

Avon and Somerset Constabulary Incident Report/s

No information identified to date and/or still to be located.

Avon Ambulance Service (now South West Ambulance Service) Incident Report

No information identified to date and/or still to be located.

Other information sources

No information identified to date and/or still to be located.

Service learning material

Further information hoping to be identified and still to be located.

Videos available

Blackdog TV (originally from ITN). (2014). UK News: Female Firefighter Killed (1996). [online].  [Accessed 13th June 2017].

Go back