Consequences of the Accident and Lessons Learned - Buncefield Fire 2005 - Case Study Example

BUNCEFIELD FIRE 2005 – ENGLAND 1. Analysis of the case study Mineral oils had been stored at Buncefield since 1968. The affected storage depot, Hertfordshire Oil Storage Terminal Limited or HOSL was the fifth biggest complex of mineral oil storage tanks in the United Kingdom. It has a capacity of 273 millions of litres of fuel. Prior to the disaster, the plant handled approximately 2.37 million tones of fuel annually. Four hundred tanker trucks daily delivered the fuels stored in Buncefield. The Health and Safety Executive or HSE performed investigations into the accidents, in collaboration with the Environment Agency or EA. The supervision committee for the investigations is the Major Incident Investigation Board or MIIB, which consist of independent experts, as well as HSE and EA employees. On December 11, 2005, several explosions were heard coming from the Hertfordshire Oil Storage or better known as the Buncefield Oil Depot. It is a major fuel storage and distribution depot located near Hemel Hempstead, some 40km north-west of central London. At six in the morning of that day, the massive explosion and the subsequent fire destroyed over 20 tanks in seven separate bunds. The fire also destroyed the large office that is almost 100m away and beyond the site boundary. The blaze lasted for five days and plume of black some from burning fuel went high into the atmosphere and was visible from several miles away. Several staff was injured, destroyed the firewater pumps and severely damaged the site offices, as there was no effective fire fighting capability available from within the site. The subsequent investigation revealed that the initial loss of primary containment was the overfilling of a petrol storage tank ‘T912’, one of three tanks in a concrete bund. This continued for approximately half an hour and several tonnes of petrol cascaded down the outside of the tank creating a large explosive vapour cloud. As the tanks became damaged, other mineral oil products were released and this led to two other explosions and a huge fire. The explosion produced overpressures that were much higher than expected and research is being carried out to understand why. It caused considerable damage to tanks and surrounding buildings. According to the final report of the Major Incident Investigation Board (2008, p.7), at about 5:30am, delivery of unleaded petrol started to arrive at Tank 912 in bund A. To prevent overfilling, a safety system was installed to shut-off the supply of petrol to the tank. However, this safety system failed and petrol cascaded down the side of the tank. Tank 912 was equipped with, among other things, devices for measuring the filling level and the temperature of the liquid inside the tank. Like all other tanks at this plant, the devices in Tank 912 were linked to an automatic tank measuring system or ATG. The levels in the tank were also monitored from a control room. Based on the monitoring data available, it could be noticed that that the tank continue to fill, via pipeline, even though the filling level measurement already showed that 2/3 of the maximum level had been reached at around 3:00 am. Moreover, in addition to the filling level measuring device, the same tank was equipped with an independent overfill safety device or visual and audible ‘End-Peak-Alert’, as well as safety switches which was meant to alert the supervisor in the control room as soon as the maximum filling level was reached. This alert was also meant to initiate the closure valves on the active pipelines. These alarm signals of the HOSL West tanks were normally also sent to monitoring and protocol system, linked to the pipelines. After the investigations, it was ascertained that no ‘End-Peak-Alert’ had been received. The personnel did not respond to the obvious failure of the filling level measuring device. The device had not displayed any changes in the filling level for over 3 hours, even though the tank had kept on filling. The personnel did not notice the leaking of substances, which lasted for about 40 minutes, either. Apparently, preliminary planning was insufficient as well, since the personnel did not know how much time it took to completely fill the tank. As the fuel continue to flow, the mixture of petrol and air formed vapour clouds over the bund wall. About 300 tonnes of petrol flowed out of the tank and approximately 10 percent of which turned to vapour that mixed with cold air eventually reaching concentrations capable of supporting combustion. The evaluation of the available information showed that the cloud stretched over an area of more than 80,000 square metres with a depth of between 1 and 7 metres, and looked like a fog. The basic cause as reported, the explosions that caused the extensive damage on and off site and the subsequent fires, can be attributed to the ignition of a flammable mixture that is mistaken as a fog. The main explosion took place at around 6:01 a.m. and measured 2.4 on the Richter Scale according to the British Geological Service. After the evaluation of the accident, it was noticed that the pressure of the explosion was 10 times higher that the preliminary expectations and calculations. The over pressure is situate somewhere around 700 to 1000 mbar on the parking lots nearby, where buildings were seriously damaged. (Moore and Lakha 2006, p.171). The exact nature of the mist is not known but experts believed that it may be an unstable portion of the original fuel or ice particles because of the mixing of air and escaped fuel. The incident confused some investigators since by nature, petrol will not easily exploded and the formation and distribution of vapour around and beyond the site is perplexing. However, the final report concluded that the most possible source of ignition is generator cabin on the south side of the Northgate building and the pump house close to the north side of bund A. There was evidence to support the view that they both located within a fuel-rich part of the vapour cloud (Major Incident Investigation Board 2008, p.12). The incident involves more than a thousand fire fighters and police forces from all over the United Kingdom. The blaze took 32 hours to extinguish and it consumed around 750,000 litres of foam concentrate and 55 million litres of water. During the initial stages of the fire, many of the bunds performed well. They contained both leaking fuel and firewater, which allowed the fire service to operate close to the burning tanks and reduced the escalation of the fire. However, loss of integrity of some bund walls developed over the following days, which allow fuel and contaminated firewater to flow out over the site (Institution of Chemical Engineers 2008, p.122). 2. Consequences of the Accident and Lessons Learned Although nobody was killed but over 40 people were injured. The explosion caused significant off-site damage to industrial and domestic properties and the fire destroyed one major off-site office block. Due to explosions and the high pressure, not only the tanks nearby, but also the stationary and half stationary fire fighting devices and the pump house, with the fire fighting pumps, were damages severely, so that they can be used for fire fighting. At the moment of the overfilling, the secondary containment assured the retention of the leaking substances. Due to the explosion, the damage to neighbouring tanks and the delay of the fire fighters, these secondary containments had to withstand a long heat impact resulting to joint sealant failures. Several houses near the depot were destroyed and some suffered sever structural damage. Residents and businesses had to find alternative accommodation while their properties were repaired. Some people lost personal possession and some were severely affected by trauma and had to be hospitalized for psychological help. Approximately 2000 residents of the nearby community were evacuated and nearby section of the M1 motorway, schools located in Hertfordshire, Buckinghamshire, and Bedfordshire were closed for a couple of days. Businesses on the Maylands Industrial Estate were also badly affected. Some companies went into liquidation and some jobs had to be relocated. The East of England Development Agency reported that approximately £70million was lost for local businesses. Civil damages claims in excess of £660 million have been lodged. The tanks and bunds involved in the fire were destroyed, along with the offices and road tanker loading bays. They were all demolished and the site cleared down to ground level. Consequently, the loss of the depot resulted in temporary disruption of fuel supplies in the southeast even though there were immediate fallback arrangements made. The site is covered with layer of clay soil approximately 5m deep, below which is a chalk aquifer used to supply potable water for the surrounding region. A number of boreholes have been sunk around the site and many groundwater samples have been taken in an attempt to establish the extent and severity of the pollution. The result suggests that groundwater under the site and up to 2km to the north, east and southeast has been contaminated with hydrocarbons and PFOS (Institution of Chemical Engineers 2008, p.124). However, this contamination was mostly near the depot and fortunately did not affect the potable water supply. The overall report by the Department for Environment on pollutants emitted from the Buncefield fires concluded that the incident did not severely affect the quality of air at ground level. Although it is quite encouraging that initial investigation concluded that the Buncefield incident did not severely affect the quality of air at ground level, there are uncertainties that the pollution may occur somewhere else. The impact of disaster on human health varies widely according to different factors such as the nature of the disaster itself, population density, pre-disaster health and nutritional status, climate, and the organization of health services. Pollutants have considerable physical and social cost and these were reflected in the impact of pollutants on plant and animal life and the associated degradation of the environments. Social cost included increased health care cost associated with higher incidences of disease, urban blight, and so on (Noji 1997, p.14). The cost of rebuilding the Buncefield depot have been estimated to be approximately £70 million and the cost the government of the investigation response is about £15 million. Some of the ’unquantified’ effects are close of motorways, loss of goods in local warehouses, momentary loss of engineering and certification services affecting the services and manufacturing sectors, momentary outsources payroll service disruptions, temporary loss of London congestion charging administrations, and temporary loss of other public service records (Major Incident Investigation Board 2008, p.30). Primarily, the harsh lessons learned from the Buncefield incident is about the vulnerability of this type of storage depot (Jones 2008, p.239). The incident has highlighted the dangerous nature and shortcomings of storing hydrocarbon fuels (Harper 2007, p.137). Moreover, the explosive properties of fuel did not seem to worry the planners who allowed the Buncefield Oil Depot to be built just 150 yards from the nearest home, on the fringe of Hemel Hempstead in Hertfordshire. The investigation clearly stated that the incident “confirms the overriding need to ensure the integrity of the primary means of containment” (Major Incident Investigation Board 2008, p.43). In short, the existing means of containment is deficient and inadequate and this should be a lesson to various regulating agencies. Another is the fact the current assessment of safety integrity did not take into account the “existence of nearby sensitive resources or populations” (Major Incident Investigation Board 2008, p.44) as the board asked the CA and operators to develop and agree to a common methodology to determine the safety integrity level requirements. The Buncefield Oil depot incident is one useful example of the need for contingency planning. The Buncefield incident provides more messages and lessons because it just happened out of the blue with no obvious reason, warning or prior experience to suggest it might. It happened at the most inconvenient time, two weeks before Christmas when demand for oil products was at its peak and many organizations were working at full capacity. What had happened were things that many had not planned for that caused the most concern, for instance, the police and other emergency service access restrictions. Lack of access to their own business premises caused anger, conflict and major frustration. The Buncefield disaster had a heavy impact on surrounding business. According to Spedding (2008, p.184), Internet retailer ASOS suspended trading of their shares, the headquarters of software firm Northgate Information Solutions and other were severely damaged and had to initiate its contingency plan. After the Buncefield disaster, several companies went bankrupt which were not even damaged by the explosion but were just in the exclusion zone for a week. This illustrates the importance of having some kind of business continuity plan in place, as well as some kind of insurance, which covers such eventualities. On the positive side, the response to the Buncefield incident demonstrates how the existence of an emergency response plan and training and coordination of first responders can lead to a rapid and successful response. The Buncefield incident suggests that the current system of safety regulation could be improved as the incident could have resulted in a substantial number of deaths. 3. Assessment of the implications if such accident happen to Preston City, England From the Buncefield incident, we learned that technology entailed social and environmental impacts that many now regard as deeply unfortunate for the recipient societies. Policy choices for managing technology may also produce unintended negative consequences. As local communities come to grips with increased vulnerabilities, they enter into new relationships with both the environment and larger social contexts, inevitably affecting the pace of social and cultural change. Furthermore, in coping with disaster impacts, communities are forced to adjust past structures and practices to altered circumstances, if only in novel forms of resistance to disaster induced changes. Any discussion of a disaster and its effects on a community must consider the issues of shorter-term organization changes and longer-term structural adaptations involving the future well-being of the community as well as the trauma of impact (Quarantelli 1998, p.232). In Buncefield, negligence on the part of both regulating authorities and operators were apparent. With all the potential disaster that Buncefield may bring, it is located in a residential and business area. What had happened in Buncefield is unique and unexpected, opening our eyes to the reality that disaster can happen anywhere. Preston is a City with a large population and thriving businesses and if such incident happens, more properties and lives will be lost. The effects on the economy would be greater and disruption of businesses and services would be higher and longer. However, Preston City is not Hertfordshire in terms of regulations. Let us remember that in Hertfordshire, a potential disaster source like Buncefield was actually allowed to be built near a community of people. The reality of Buncefield and the lessons learned makes it improbable that a certain community, who value the life of its members, would make the same mistake and tolerate such potential hazardous infrastructure to be built near them. 4. Bibliography Harper Gavin D. J. 2008. Fuel Cell Projects for the Evil Genius. McGraw-Hill Professional, US Institution of Chemical Engineers. 2008. Hazards XX: Process Safety and Environmental Protection, Harnessing Knowledge, Challenging Complacency. Institution of Chemical Engineers (IChemE), UK Jones Conrad. 2008. Soft Target. AuthorHouse, US Major Incident Investigation Board. 2008. The Final Report of the Major Incident Investigation Board. The Office of the Public Sector Information, UK Moore, Tony and Lakha Raj. 2006. Tolley's Handbook of Disaster and Emergency Management: Principles and Practice. Butterworth-Heinemann, UK Noji Eric K. 1997. The Public Health Consequences of Disasters. Oxford University Press, US Quarantelli Enrico Louis. 1998. What is a Disaster?: Perspectives on the Question. Routledge, UK Spedding Linda. 2008. Due Diligence Handbook: Corporate Governance, Risk Management and Business Planning. Butterworth-Heinemann, Hungary Read More