Resources » Buncefield Fire (GB)

On the 11th December 2005 at approx. 6:00 hours (UTC), several explosions were registered at the Buncefield oil storage terminal nearby the town of Hemel Hempstead within the county of Hertfordshire to the north of London.

The Buncefield oil storage terminal is situated near (approx. 15 km) the airports London Heathrow and Luton. Due to the convenient location, besides fuel oil, gasoline and kerosine are stored and distributed as well. The total mass is comprised of approximately 150'000 tons stored within 26 tanks. The fuels arrive at Bruncefield via several pipelines. Delivery is achieved using fuel tanks. The frequency is around 400 fuel tanks per day (see Total Homepage).    

The Explosions and the Fire

On the 11th December, 3 explosions were registered. The pressure blast caused by the explosion was perceptible up to a distance of 100 miles (about 160 km).  Glass could be observed shattering up to a distance of 2 miles (3.2 km) (refer to RSC Adavancing the Chemical Sciences). Columns of fire reached a height of over 60 metres. More than 43 persons were injured in the explosion, 2 of them seriously (refer to BBC News). Approx. 2000 persons were evacuated. A total of 20 tanks caught fire.

Image produced from the Ordnance Survey Get-a-map service. Image reproduced with kind permission of Ordnance Survey and Ordnance Survey of Northern Ireland.
Satellite Photo showing the extent of the cloud of smoke produced by the fire. Source: NOAA (National Oceanic and Atmospheric Administration), USA

Comparison with Crude Oil Fires during the first Gulf War

During the first Gulf War (1990/1991) 700 crude oil sources were ignited, beginning as of the 23.2.1991 in Kuwait. The fires were still being extinguished up until the beginning of November 1991.

Measurements have shown that approx. 93% [Cofer et al, 1992] and 96% [Ferek R.J et al, 1992] respectively of the carbon is converted to carbon dioxide. The fraction of carbon monoxide was in the range of 0 up to 1.5%. That is evidence that the combustion proceeded quite completely, when you compare this to fires in buildings where, as a rule of thumb, the carbon monoxide fraction makes up approx. 20% of the carbon converted [Gann R.G at al, 2003].

Kuwait:  Measurements taken from an Aeroplane in the Clouds of Smoke produced by the Fires 

Concentration measurements of carbon dioxide, carbon monoxide, methane, sulphur dioxide, nitrogen dioxide and hydrogen sulphide, taken directly in various clouds of smoke by an aeroplane, did not exceed the  ERPG-2 and TEEL-2 toxicity values respectively (refer to the following table [Ferek R.J et al, 1992]):

Gas Smallest
average concentration
average concentration
ERPG-1 oder
ERPG-2 oder
Carbon dioxide 33'000  132'000  54'000'000 
Carbon monoxide 21  616  229'000  400'000 
Methane 17  132  196'800  3'280'000 
Sulphur dioxide 40  1300  800  8000 
Nitrogen dioxide 29  1'200  18'400 
Hydrogen sulphide    140  42'000 
Ultrafine respirable dust<3.5µm   64  1'093 
Soot/Carbon particulate matter  308 

The measurements in the above table show that the toxicity value ERPG-1 is only exceeded by sulphur dioxide. By the way in our MET model the ERPG-1 value is used as a threshold value, i.e. all concentrations which fall below this this concentration are not taken into consideration.

In the same way, traces of acetylene, butane, ethane, ethylene, heptane, hexane, propane, propane, propylene, benzene, toluene, xylene were measured in the gas cloud. Further gases which form due to photochemical reactions, like formaldehyde and ozone, were also measured [Luke W.T. et al, 1992]. All of these gases remained below the corresponding ERPG-2 values. The ERPG-1 concentrations were exceeded by ozone.

Kuwait: Drop in Sulphur- and Nitrogen-Oxide Concentration

The sulphur concentration in Kuwaiti crude oil is approx. 2 to 3 Wt %. Measurements taken from an aeroplane, however, showed that the average sulphur concentration due to sulphur dioxide and sulphate within the gas cloud was only about 0.5% [Cofer et al, 1992]. Measurements taken directly at the source, however, have yielded a total sulphur concentration of approx. 2.2%, which corresponds well with the sulphur concentration in crude oil. After further investigations, a decline in sulphur dioxide over time could be proved which is about 6% per hour. A decline of 22% per hour could be measured for nitrogen oxides. Elution due to rain-out could not have accounted for the decline, because the  relative air humidity was less than 20% [Ferek et al, 1992].

Kuwait: Comparison of the Results measured on the ground with the Immission Threshold Values 

Measurements taken at ground level of toxic substance concentrations have also been carried out by various organizations. Some gases and the EU and CH immission threshold values are listed in the following table:

Substance Concentration [µg/m3] Immissions Threshold values [µg/m3]
- Average Max. EU Meaning in the EU CH Swiss Meaning (in CH)
Carbon monoxide
Ref 1
800 51'300 10'000 As the heighest 8-hour-long averaged value measured during the course of one day 8'000 24-h-average value; may only be exceeded for a maximum of once per year
Carbon monoxide
Ref 2
1'058 8'576 10'000 As the heighest 8-hour-long averaged value measured during the course of one day 8'000 24-h-average value; may only be exceeded for a maximum of once per year
Ozone 53.4 104.8 120.0 As the highest 8-hour-long averaged value measured during the course of one day, may be exceeded for a period of 25 days of a calendar year, averaged over 2 years 100 98 % the ½-h-mean values measured for one month
Sulphur dioxide 23.8 92.8 80.0 Median of the daily average values taken over a period of a year, assigned >40 µg/m3 dust median of the TMW per year 80.0 95 % the ½-h-mean values measured for one year
Nitrogen oxides 58.5 86.1 40.0 Immission threshold value (health) Average value taken over one calendar year as of 2010 30.0 Mean annual value (arithmetical mean value)
7.8 13.1 5.0 Mean annual value to be adhered to as of 1.1.2010 - -
Ultrafine dust 354 3000 50 More than 24 hours , maximum of 35 exceedances per year 50 24-h-average value; may be exceeded for a maximum of once per year
  • Sources relating to the table:
    Ref 1: Carbon monoxide values according to the Kuwait Environmental Protection Department Monitoring for the month of June in 1991 at the Rega measuring site. The maximum values relate to a peak value with a 5 minute duration.
  • Ref 2: Carbon monoxide values according to the "Team from France". 26.3.1991 until 6.4.1991. The maximum values relate to a peak value with a duration of 5 minutes.
  • All measurements taken from the US Army Environmental Hygiene Agency, Final Report - Kuwait Oil Fire Risk Assessment, No. 39-26-1192, February 1994
  • EU Immission threshold value according to the 'Compilation of Immission Values' of the Federal Environmental Agency in Dessau.
  • CH Immission threshold according to the LRV (the Swiss Clean Air Maintenance Regulation called the ‘Luftreinhalte-Verordnung’) 814.318.142.1
Photos: Jack Heller, USACHPPM, May 1991

A comparison of the average results measured with the immission threshold values shows that the average concentration of nitrogen oxides, benzene and PM10 ultrafine dust has exceeded the immission threshold values.

Indeed these values (with the exception of ultrafine dust) are comparable to the values measured in US suburbs and are generally smaller than those of large US cities [USDD, 1994].

Kuwait: PM10 Ultrafine dust concentration

The PM 10 ultrafine dust concentration is relatively high in Kuwait or Saudi Arabia due to natural emissions (from the desert) in comparison to European conditions with an immission threshold value of 50 µm/m3 . The average PM10 concentration almost reaches a level of 600 µm/m3 in certain areas. PM 10 values measured during May and October 1991 are listed in the following table [USDD, 1994]:

Ort Average 24h PM10 Concentration [µg/m3] Max 24h PM10 Concentration [µg/m3]
Ahmadi Hospital, Kuwait 367.5 544.1
Al Eskan Village 271.1 697.5
King Khalid Military City 298.6 923.5
Khobar Towers, Saudi Arabien 267.8 433.8
US Botschaft, Kuwait 670.6 1105.7

The analysis of the chemical composition of the ultrafine dust showed that the dust in Kuwait was composed of approx. 23% carbon and up to 75% of silica and calcium compounds. In Saudi Arabia where the influence of the fire gases was minimal, the carbon component was less than 5%.

Are the results of the first gulf war fires transferable to the Buncefield Fire Case?

In Kuwait crude oil and methane were burned, in Buncefield fuel oil, gasoline and kerosene. Due to legal regulations, the sulphur content of the fuels in the Buncefield fire should therefore be lower than that of the crude oil from Kuwait. In Switzerland, the maximum allowable sulphur content in benzene about 50 mg/kg. The sulphur dioxide output should therefore be far smaller per kilogram of burned fuel. The quantity of fuels burned in the Buncefield case was far smaller and the fire only lasted for a few days, whereas in Kuwait the fires lasted for several months. Combustion products which occur due to the combustion of additives may have produced different compounds in the case of Buncefield.

Nevertheless, the following interpretations are, without any doubt, applicable to both cases:  

  • The combustion of carbon to carbon dioxide in open crude oil fires is high in comparison to other fires involving biomass e.g. wood fires. The carbon monoxide component is therefore relatively small.
  • The acutely toxic concentration value ERPG-2 was not exceeded by any of the fire gas components measured in Kuwait.  The ERPG-1 value was only exceeded by sulphur dioxide and ozone.
  • Ground measurements did not yield any concentrations which exceeded the acutely toxic ERPG-2 or ERPG-1 values. The immission threshold values were partly exceeded for benzene, but were within the range of values typical for large cities [USDD, 1994]. 
  • The ultrafine dust load in particular, which nevertheless has no acutely toxic effect, becomes problematic during a crude oil fire.
  • The estimation of critical distances for the toxicity of a crude oil fire therefore do not make any sense, because the threshold concentration (ERPG-1) even within the gas cloud itself produced by the fire, lies below this threshold for most of the components

  • Biever C, Lucky escape after fuel depot explodes, New Scientist 188 (2005) 17
  • Ferek R.J, Hobbs P.V., Herring J.A., Laursen K.K., Weiss R.E, Rasmussen R.A, Chemical Composition of Emissions from the Kuwait Oil Fires, Journal of Geophysical Research, 97 (1992) 14483-14489
  • Gann R.G, Averill J.D, Johnsson E.L., Nyden M.R, Peacock R.D, Smoke Component Yields from Room-scale Fire Tests, National Institute of Standards and Technology Technical Note 1453 (2003), Gaithersburg USA
  • Luke W.T., Kok G.L., Schillawski R.F., Zimmerman P.R., Greenbarg J.P., Kadavanichi M, Trace Gas Measurements in the Kuwait Oil Fire Smoke Plume, Journal of Geophysical Research, 97 (1992) 14499-14506
  • US Army Environmental Hygiene Agency, Final Report - Kuwait Oil Fire Risk Assessment, No. 39-26-1192, February 1994
  • US Department of Defense, "Report of the Defense Science Board Task Force on Persian Gulf War Health Effects," Washington, D.C., June 1994

no news in this list.