Fourth International Symposium on Tunnel Safety and Security, Frankfurt am Main, Germany, March 17-19, 2010

Tunnel Fire Safety in Australasia

Arnold Dix Scientist and Lawyer Associate Professor, UWS, Australia

ABSTRACT Fixed fire fighting systems are used in all high risk urban tunnels in Japan and Australia. No tunnel fitted with such systems has experienced a catastrophic fire over the last forty years. The 2007 Burnley fire in Australia provides a rare insight into how fires are kept small by operation of such systems. KEYWORDS: Fixed fire suppression system; Deluge; Burnley; Japan. INTRODUCTION Fixed fire suppression systems have been successfully used for more than 40 years in Japans congested urban road tunnels and more recently in all of Australias congested urban tunnels. There have been no deaths attributable to tunnel fires and no significant infrastructure damage following any of the numerous crashes and fires in any tunnels fitted with the Japanese or Australian designed integrated fixed fire suppression systems. This is in contrast to other parts of the world which do not routinely use fixed fire fighting systems. Other incidents including the Mont Blanc, Tauern, Gotthard, and Frjus Tunnel fires, all began with comparatively small events but without rapid intervention from a fixed fire fighting system the fires escalated causing significant loss off life and infrastructure damage. The fatal Burnley Tunnel incident in Australia of 23 March 2007 provides a rare insight into the effectiveness of these fixed fire fighting systems. Unfortunately most of the technical details of the Burnley Incident remain secret. The detailed investigation report is subject to a legal suppression order and the Report to the Victorian Coroner, The Fatal Burnley Tunnel Crashes Melbourne, Victoria, Australia (Arnold Dix 2008) cannot be made public. This paper uses evidence from the Supreme Court criminal proceedings of mid 2009 which is not suppressed to reveal some of the events of 2007. Fixed fire fighting systems are a proven tool for protecting both human life and tunnel infrastructure. When used intelligently, by competent operators, as part of a rapid and co-ordinated response to an incident, the Burnley incident demonstrates fixed fire fighting systems are effective in protecting tunnel infrastructure and delivering human safety. Although it must be conceded that the statistical significance of the Japanese and Australian experience is unclear, neither Japan nor Australia has experienced a catastrophic tunnel fire despite numerous tunnel fire incidents. The Burnley fire provides a critical insight into how fires in Japanese and Australian tunnels are managed. The challenge for Europe is to translate the positive Asian experience of fixed fire fighting systems into an approach suited to European conditions. Europes grand tunnelling achievements of the 19th and early 20th century deserve due regard to the learning and experience from the exotic Far East.

69

Fourth International Symposium on Tunnel Safety and Security, Frankfurt am Main, Germany, March 17-19, 2010

Background to Fire Suppression Tunnel fire incidents have demonstrated loss of life in circumstances which were not previously considered high risk scenarios. Many of the factors identified as central to the extensive loss of life and damage in other events were shared with the Burnley event yet the consequences were by comparison minor. The 5 unifying themes to emerge from the analysis of the catastrophic international events are: Fires develop much more quickly than expected; Fire temperatures of in excess of 1000 C are achieved; Smoke volumes are higher than expected from an early stage of the fire growth; Fire spread between vehicles occurs over a much greater distance than had been expected previously (e.g. more than 200 m in the Mont Blanc Tunnel); The road tunnel users behaved unexpectedly, they: Did not realise the danger to which they were exposed; Failed to use the safety infrastructure provided for self rescue; Believed (wrongly) that they were safer in their cars than if they used the self rescue safety systems; Chose to stay in their vehicles during the early stages of a fire as they did not want to leave their property; Realised too late the danger they had placed themselves in, by which time it was too late to execute self rescue.

Yet in the Burnley incident, despite immediate explosions and a large fire involving trucks the consequential damages were, by comparison, small. These observations indicate that there are a range of opportunities to improve tunnel safety following a fire incident and that there may be lessons about what went well in the Burnley incident. The single most likely critical factor is the use of an effective fixed fire fighting system which was successfully integrated into other safety systems. THE BURNLEY INCIDENT The Burnley tunnel is a single direction, 3 lane tunnel of 2.9km length. It has a traffic flow of around 100,000 vehicles per day. On 23 March 2007, at 09:52:30 am a truck travelling Eastbound made an unscheduled stop in Melbournes CityLink Burnley tunnel. Over the next two minutes 103 vehicles passed the stoped truck without incident. Two minutes later, by 09:54:24 seconds several vehicles, including 4 HGVs and 7 light vehicles had crashed, 3 people were dead and fire and a series of explosions were initiated. By 09:56:00am (two minutes after ignition) emergency ventilation and a fixed fire suppression system had been activated. The following eye witness evidence from the criminal court case against the driver graphically illustrates the sequence of events: o o Just describe in detail for the jury what exactly you saw? ---Essentially I heard the screeching of tyres. I looked in the rear vision mirror; saw the car careering into the back of the truck. The nose of the car went down, the car lifted up like that so - and then there was another smash from behind by a truck. So the smash from behind from the truck, what did you actually see?

70

Fourth International Symposium on Tunnel Safety and Security, Frankfurt am Main, Germany, March 17-19, 2010

o o o o o

---I saw the truck hit it and the - I can only assume that it was the gas tank of the vehicle that exploded. So you heard an explosion? ---I saw the explosion. You saw it? ---Yes. So you saw the truck hit the car and then there's the stationary truck in front of it, so the car is effectively between the two trucks? ---Essentially, yes. You've told us you've heard an explosion. Was it a loud explosion? ---It was a reasonably loud explosion, yes. Did you continue to drive through the tunnel or what did you? ---Well I continued to drive through the tunnel. There were - there was another explosion shortly after that which was a much bigger explosion. I remember the windows of my car vibrating as a result. There was also another announcement that came over the speaker saying that there had now been an incident in the tunnel and that vehicles were to slow down to 60 kilometres an hour

This evidence graphically describes the crash, fires and subsequent explosions. The initiating events for this incident were large large in the context of prior catastrophic events such as the engine compartment fire at Mont Blanc. Yet, despite the severity of these initiating events the fires were contained, with no flash over or other significant fire growth occurring once the deluge fixed fire fighting system was initiated. However, it was not merely the presence of the fixed fire fighting system which was critical it was that the ventilation system was effective and that the system was operated in a timely and accurate manner. Furthermore it was the fire brigade that put the fires out the deluge system merely kept the fires small enough to allow effective emergency services intervention. Following is a summary of key events for the 2007 Burnley Tunnel Fires as given in evidence in the Supreme Court Proceedings DIRECTOR OF PUBLIC PROSECUTIONS v. DAVID LAWRENCE KALWIG 16/07/2009 [This evidence is indicative only of the events the currently suppressed Technical Report for the Coroner (Dix 2008)] contains detailed technical information on the events): At 9:52:30 a truck stopped in the left (slow) of three lanes in the east bound Burnley Tunnel. At 9:52:34 the tunnel operator acknowledged a stopped vehicle alarm At 9:54:10 the tunnel operator commenced a lane closure and speed reduction plan. From 9:54:16 and 9:54:22 Computer controlled lane closure and speed signs changed to reflect the left lane closure and speed reductions in the tunnel. At the same time the left lane closure was being implemented and the speed reductions and radio rebroadcast messages were being played a group of trucks and other vehicles approach the area of the tunnel where the truck had stopped in the left lane. At 9:54:22 a group of vehicles were slowing in all lanes (nearly stopped as they approach the area of the tunnel of the stopped truck)

71

Fourth International Symposium on Tunnel Safety and Security, Frankfurt am Main, Germany, March 17-19, 2010

There was nothing exceptional about the movements of the stopping light vehicles before the incident. In all instances, each was observed to be preparing to negotiate the stopped vehicle in a responsible and courteous manner, At approximately the same time, a HGV truck travelling in the left lane (much more quickly than all other vehicles in this region of the tunnel) changed lanes rapidly from the left inside lane into the centre (and partially right lane) initiating a series of collisions directly impacting 5 cars and three other semi-trailers. Between 9:54:26 and 9:54:30 a series of collisions, explosions and fires occur in the region immediately behind the stopped truck. Eventually the truck which initiated the series of crashes hit the stopped truck and pushing it many metres forward. Between 9:54:26 and 9:54:33 other tunnel users, not directly involved in the crashes conducted emergency breaking manoeuvres and successfully brought their vehicles to a stand without further collisions occurring. This group of users were travelling near or behind the crashes. People were evacuated from 9:54:42. Between 9:54:29 and 9:55:14 the Burnley Tunnel Controller Room Operators initiated an emergency response which included the closure of the Burnley Tunnel. From 9:55:12 radio rebroadcast messages were transmitted in the Burnley Tunnel indicating a left lane closure and a speed reduction. At 9:55:12 CityLink initiated break glass alarm to Fire Brigade. At 9:55:15 the Operator closed the Burnley Tunnel. At 9:55:37 the tunnel operator enabled emergency mode in preparation for the smoke extraction, deluge operation and evacuation. At 9:55:50 the emergency response plan was initiated by the tunnel controller including activation of emergency smoke extraction and the deluge system. At 9:55:54 the smoke extraction system was activated. At 9:56 the fixed fire suppression (Deluge) was activated. There was an unexpected delay of approximately 30 seconds for one of the deluge zones to activate At 9:57:22 evacuation mode was implemented to evacuate users via the domain tunnel.

Three people were killed in three different vehicles. Two of the three deaths were determined to be effects of fire. The fires which killed these people were not, and could not, be extinguished by the deluge system. All those killed suffered serious physical injuries in the car crashes. The incident resulted in several hundred people being evacuated from the tunnel and their vehicles. None of the evacuees or their vehicles was injured or damaged. The tunnel only suffered minor damage, and could have been re-opened 10s of hours later if the extent of the damage could have been more rapidly determined.

72

Fourth International Symposium on Tunnel Safety and Security, Frankfurt am Main, Germany, March 17-19, 2010

Mont Blanc Tunnel after fire (no suppression system.)

Burnley Tunnel after fire (with suppression system). NB road surface and tunnel walls and services still intact.

These outcomes from the Burnley incident are entirely consistent with the general experience in Japan of such systems. So long as the ventilation and fire suppression systems are activated in a timely manner and in the correct mode it is the experience in Japan and Australia that the fire event size will be contained, fire spread controlled and a potential catastrophe avoided. It is this reasoning which underlies the continued investment in such systems in Japan and Australia. It is acknowledged that if an initial event is too large these systems will not provide infrastructure or personal life safety protection. BACKGROUND FIXED FIRE FIGHTING SYSTEM World Context Fixed fire fighting systems have routinely been installed in Japanese tunnels for over 40 years. For all highly trafficked urban tunnels they are prescribed. In the United States of America they are installed in only three tunnels and in Australia they are installed in all urban road tunnels. The use of such systems is the exception in all countries except Australia and Japan. PIARC on Fixed Fire Fighting Systems a Short History PIARC has changed its position on the installation of fixed fire fighting systems from not recommending to recognised for asset and life protection from 1983 2008. In PIARCs report to the Brussels World Road Congress in 1987 PIARC concluded that: Given our present knowledge of the subject the use of sprinkler systems in road tunnels is not recommended for the following reasons: The water can result in the dispersion of burning liquids over a large surface area Water can cause dangerous reactions when it comes into contact with certain products Even if the flames are extinguished the metal parts of the vehicles do not cool very quickly

73

Fourth International Symposium on Tunnel Safety and Security, Frankfurt am Main, Germany, March 17-19, 2010

Flammable products can continue to give off gases leading to the presence of explosive mixture The water (or foam) that is distributed may, as in the case of the Velsen Tunnel in Holland, be insufficient to extinguish the fire on a vehicles carrying a large quantity of fuel At the Montreal World Congress in 1995 PIARC again concluded that: Sprinkler systems are not recommended in road tunnels. In 1999 by way of the publication, Fire and smoke control in road tunnels PIARCs position softened. No European countries use sprinklers on a regular basis. For some tunnels in Europe sprinklers have been used for special purposes. In Japan sprinklers are used in tunnels longer than 1,000 metres to cool down vehicles on fire. In the United States only tunnels carrying hazardous cargo may have some form of sprinkler. The reasons most countries do not use sprinklers in tunnels, is that most fires start in the motor room or in the compartment, and sprinklers are of no use till the fire is open. Sprinklers can be used, however, to cool down vehicles, to stop the fire from spreading to other vehicles (i.e., to diminish the fire area and property damage) and to stop secondary fires in lining materials. Experiences from Japan show that sprinklers are effective in cooling down the area round the fire, so that fire fighting can be more effective. In 1999 PIARC summarized the problems with sprinklers as follows: water can cause explosion in petrol and other chemical substances if not combined with for example Aqueous Film Forming Foam (AFFF); vaporized steam can cause damage to people; the efficiency is low for fires inside vehicles; the smoke layer is cooled down and may cover the whole tunnel; maintenance can be costly; sprinklers are difficult to handle manually; sprinklers should not be used in the fire area before all people are evacuated; automatic sprinklers can pose difficulties for fire fighting. Accordingly PIARC eased it positions concluding that: the working group will normally not recommend sprinklers for ordinary fire fighting. Sprinklers in tunnels are primarily used to protect the tunnel structure not the motorists. Sprinklers can also be used, where appropriate, in ancillary rooms in tunnels and tunnel facilities. In 2007 PIARC, in its publication, Systems and equipment for fire and smoke control in road tunnels, which was actually written in 2005 PIARC concluded: at the moment, an owner/operator who wants to install new detection and new fire fighting measures must properly verify that the conditions for installing, using and maintaining contribute to the overall safety and are compatible with the framework of the entire safety concept for that specific tunnel. He must also ensure the effectiveness of the proposed measures. PIARC currently acknowledges that such systems can be included in tunnels where they form part of the total fire safety concept for a tunnel and that a systems engineering approach is taken to their integration.

74

Fourth International Symposium on Tunnel Safety and Security, Frankfurt am Main, Germany, March 17-19, 2010

NFPA 502 NFPA 502 Standard for Road Tunnels, Bridges and Other Limited Access Highways has in its 2008 addition accepted that fixed fire fighting systems can be effective in controlling vehicle fires in road tunnels by limiting the spread of the fire, reducing the temperature of surrounding air and surfaces, and reducing the fire heat release rate. (Over View of NFPA 502 Standard for Road Tunnels, Bridges and Other Limited Access Highways, 2008 Edition - William Connell and Jason Gamache 2007.) Like PIARC, NFPA 502 had resisted endorsing the use of such systems for reasons including: the systems inability to extinguish fires within motor components the production of super heated steam destratification of toxic gas and smoke risk of tunnel users thinking spraying water was as a result of tunnel structural failure the costs of maintenance. Each of these concerns have now been addressed by the standard fixed fire fighting systems are therefore acknowledged as a potentially valuable tool in protecting human life and assets. SIGNIFICANCE OF INTERNATIONAL CHANGES These changes to standards squarely raise the potential for fixed fire fighting systems to be used responsibly in road tunnels throughout the Western World. From a corporate governance and professional negligence perspective failure to consider the use of such systems may constitute evidence of either unprofessional conduct or potentially criminal conduct in the event that fires cause fatalities and/or significant damage to infrastructure. The inclusion of these provisions does not mandate their use in a fire. There are a range of valid reasons not to use such systems and failure to explore these aspects of any decision regarding their use could likewise pose a risk in the event of their unsuccessful use in an emergency. NEW TECHNOLOGY NEW CHALLENGES The recent support for the potential use of fixed fire fighting systems by PIARC and NFPA heralds the end to a long fought battle on the need to recognise fire suppression systems as a legitimate way of managing the risks of fires in modern road tunnels. However, the challenges of such recognition are far more complex than the arguments in favour of allowing such systems to be considered. The inclusion of a fire suppression system does not of itself create safe tunnels. The maintenance of a fire suppression system does not mean that the system will operate, and the use of fire suppression does not guarantee that it will work. Such systems, if they are to be relied upon, demand an ongoing and systematic approach to their successful operation in the unlikely event of a fire. Arguments that such systems should allow a reduction in the level of fire protection and other safety systems highlight the importance that such fixed fire fighting systems actually work when an incident occurs. Such systems necessarily add to the expense of a project, require ongoing maintenance, must be operated effectively (correct location and in a timely manner) and must be integrated within other safe 75

Fourth International Symposium on Tunnel Safety and Security, Frankfurt am Main, Germany, March 17-19, 2010

tunnel safety systems. In addition, particular circumstances of each tunnel are different and those differences can be further complicated by changes in time. Accordingly, the process of determining that such a suppression system is required must be coupled with a very clear understanding of what it is that the suppression system is designed to achieve and how achieving that performance outcome will be assured. There are two main purposes for fire suppression systems. One purpose is for human protection and the second purpose is for asset protection. Protecting these two different aspects of a tunnel are not necessarily consistent. A suppression system can readily be designed to protect infrastructure but this can easily make the environment untenable for users. Examples of such systems include those using inert gases, CO2 enrichment and a range of toxic additives. There are a range of fixed fire fighting systems which vary from the Australian type deluge systems to those which include exotic fire suppressants and toxic chemicals. The purpose of the systems must be clearly understood and articulated prior to designing the system. There are a range of technical options for fixed fire fighting design on matters as diverse as droplet size, liquid spread, discharge volume and spray density to the size of extinction zones , speed of application and the temperatures in which they are to operate. At the control level there are a range of opportunities to fully integrate such systems with the ventilation, operate them separately, fully automate them, automate them with manual override, manually operate them with auto override there is a range of options as broad as the imagination of the engineers and technicians whom create them. As with all engineered systems, their actual performance is a function of both the physical environment in which they are placed and their performance specifications. Discharge densities, operating pressures and a range of other factors each contributing to a particular performance characteristic. One of the challenges with such technologies is that many clients confronted with these technologies for the first time do not have (nor could reasonably have) sufficient expertise in relation to the fire engineering which underlies their effective performance to critically review suitability of the product offered by the vendors or the technology developed by engineers working specifically on a project. TRAFFIC SUPPRESSION SYSTEM INTERFACE Suppression systems must interface with the tunnel. This point is critical because it is within the tunnel that fires and other severe incidents may occur and from which the fire suppression system must be protected. To operate effectively not only must the system have been properly maintained, the operator must be able to activate it correctly and it must survive the events which have resulted in the incident requiring its activation. Because the technology is required to work at the site of the incident, it is reasonable to expect that there is a risk that in the unlikely event that the systems are used, the crash event itself poses a threat to the system. These threats can be divided into two types. One is physical (through mechanical collision) and the second is through heat. Mechanical In many instances, the events which lead to a serious fire are as a result of a crash. In tunnels, tunnel crashes often involve the walls of the tunnel and where the ceiling is low enough could conceivably 76

Fourth International Symposium on Tunnel Safety and Security, Frankfurt am Main, Germany, March 17-19, 2010

involve the ceiling as well. Regard must be had to the risk of vehicles compromising the operation of suppression systems when the crash occurs. Ideally, this should include some form of physical protection for the suppression system along the walls of the tunnel and sufficient separation vertically to protect the suppression infrastructure from being compromised in an incident. What constitutes an appropriate treatment for protection of this physical harm of the suppression infrastructure will depend from application to application. Thermal The best means of protection for the system from thermal damage are soundly designed materials, inherent heat resistance and/or insulation and early and accurate activation. IMPORTANCE OF THE BURNLEY INCIDENT Within the context of these important changes to international standards with respect to the effectiveness of fixed fire fighting systems the Burnley Tunnel incident of March 2007 promises rare insights into the practical effectiveness of such systems. While the findings of this investigation are not public at the date of this paper critical to the analysis of the incident is a detailed understanding of the facts and circumstances surroundings the incident. In particular the use of the CityLink tunnels of advanced computer systems has allowed a second by second analysis of the events through captured video images and recorded computer logs from the many computer control systems. From this analysis not only can objective evidence on the effectiveness of the electromechanical systems to respond to an emergency readily extracted but crucial information about the tunnel controllers ability to identify an event and correctly formulate an effective response also be revealed. Ultimately forensic analysis at this level provides a tangible example of both the strengths and weaknesses of a tunnels' safety systems. Such events are a rare test for operational readiness and vulnerability for the safety systems. When the findings become public, the importance of responsibly managing the matters which are necessarily raised as necessary and incidental to the use of fixed fire fighting systems will be revealed. SYSTEMS APPROACH Fixed fire fighting systems are but one tool in a complex system designed to manage safety. Their inclusion in a project should not be seen as a formality or a routine matter. Careful formulation of their design for the specific application having regard to the peculiarities of the particular tunnel is essential but so to is ensuring that the systems will work as expected when commanded and that those entrusted with their control are sufficiently competent to operate them in a timely manner. A failure in any component of this system would render a road tunnel unsafe. Provision of a fixed fire fighting system does not of itself make a tunnel safe. Any decision to install a deluge system must be coupled with detailed technical analysis of the nozzle performance. Critical factors such as droplet size distribution, trajectory modelling of droplets through a range of longitudinal velocities is essential. An appropriate nozzle performance for the anticipated longitudinal airflow must be selected to ensure an appropriate droplet spread and mass flow performance for given water pressures.

77

Fourth International Symposium on Tunnel Safety and Security, Frankfurt am Main, Germany, March 17-19, 2010

For effective deluge operation activation must be rapid and accurate. If discharged in this way fire growth rates are likely controlled, the risk of rapid fire spread minimised and thereby toxic gas and smoke generation volumes contained. The undesirable consequences of its activation such as smoke destratification, increased humidity and decreased visibility are hopefully outweighed by their other positive outcomes of fire growth rate control, containment of fire spread, and reduced temperatures. These positive outcomes of using deluge have been consistently observed in real events in Japanese tunnels and will be the subject of expert analysis in the CityLink investigation. FIRE IN PERSPECTIVE From the material available publically in the Burnley Tunnel fires it appears the fires may have been remarkable in three important respects (preliminary observations based on public information): (a) The fires began instantaneously through the initial series of collisions generating a fire of 10s of megawatt almost immediately; (b) Despite the availability of flammable material fire growth was limited; and (c) Despite the availability of fuel (vehicles) in the tunnel, fire spread was limited. Instant Fire The immediate fire had an almost immediate heat release rate of 10s of megawatts. This was caused by the initial crashes between the truck and other vehicles. Fires of these magnitudes are extremely dangerous in tunnels. People within the tunnel are at great risk from the enormous volumes of toxic gas and smoke generated coupled with the very high rates of heat transfer in the tunnel air. Furthermore, at such high heat release rates fires jump hundreds of metres along tunnels between vehicles. (This is the subject of further discussion below). In the Burnley Tunnel incident, having a vigorous and multi-seated series of fires (several fires) which each had an ample supply of fuel, and air (with other vehicles comparatively close by) made this incident in world terms extremely dangerous. Limiting the fire growth and stopping the spread of the fire within the Burnley Tunnel was a major technical achievement. The only other country which has achieved such results is Japan. Japan and Australia share one technology in common the deluge system for fire fighting. OPERATION It is not sufficient to have an operational deluge system it is essential that it be operable. Developing an operational culture which is able to actively respond to incidents is essential for optimal use of these systems. Encouraging an operational culture which prides itself on timely and appropriate responses is an essential operational objective.

78

Fourth International Symposium on Tunnel Safety and Security, Frankfurt am Main, Germany, March 17-19, 2010

CONCLUSIONS The effective use of fixed fire fighting systems such as deluge demands a systems engineering approach to integration of the deluge system. From its initial design to its integration into emergency ventilation and alarm systems its successful operation in an event is critical to protecting the welfare of tunnel users and tunnel infrastructure. Timely and accurate activation are central to this task. Such effective use will only occur where the systems have been designed, integrated, maintained and routinely exercised. When the Burnley, CityLink investigation results become public, the ability of the Burnley tunnel operator to deliver these important outcomes in a real event will be revealed. Japan introduced fixed fire fighting systems into its high risk urban tunnels over 40 years ago. The Japanese rationale is that they minimise fire growth rates and minimise the likelihood of flash over and fire spread. Since their installation in Japan there have been no catastrophic tunnel fires in Japan. Australia has also developed fixed fire fighting systems for all its tunnels. The 2007 Burnley fire suggests the deluge system played a majoir role in protecting people and property following the crash and subsequent explosions. For high risk tunnels the question should no longer be whether fixed fire suppression should be used the question should be, Why not? REFERENCE LIST 1. 2. 3. 4. 5. 6. Technical Committee on Road Tunnels, Proceedings of the XVIIIth World Road Congress, Brussels, Belgium, 13-19 September 1987. Committee Road Tunnels. Proceedings of the XXth World Road Congress, Montreal, Canada, 3-9 September 1995. World Road Association (PIARC). Fire and Smoke Control in Road Tunnels, World Road Association (PIARC), Paris, France, 1999. World Road Association (PIARC), Systems and equipment for fire and smoke control in road tunnels, World Road Association (PIARC), Paris, France 2005. National Fire Protection Association, NFPA 502 Standard for Road Tunnels, Bridges and Other Limited Access Highways, NFPA, Qui9ncy, Massachusetts, 2008(T) W Connell; J Gamache; 2007; Over View of NFPA 502 Standard for Road Tunnels, Bridges and Other Limited Access Highways, 2008 Edition.

79

Fourth International Symposium on Tunnel Safety and Security, Frankfurt am Main, Germany, March 17-19, 2010

80