article

Analysis of transport of dangerous goods through tunnels

Posted: 13 June 2007 | J.E. Nieuwenhuis M.Sc, ARCADIS Infra BV, Amersfoort, Netherlands | No comments yet

Both ADR and RID regulations enable the competent authorities responsible to restrict the transport of dangerous goods through tunnels. However, account should be taken of the tunnel characteristics, and a risk assessment including the availability and suitability of alternative routes and modes and traffic management should be conducted.

ARCADIS has studied the transport of dangerous goods for a 53 kilometre railway tunnel through the Alps between Italy and France. Transport of dangerous goods in both train and trucks on trains is foreseen in the future. Analysis, modelling and studies about the possible effects of an accident involving dangerous substances in the tunnel were carried out.

Both ADR and RID regulations enable the competent authorities responsible to restrict the transport of dangerous goods through tunnels. However, account should be taken of the tunnel characteristics, and a risk assessment including the availability and suitability of alternative routes and modes and traffic management should be conducted. ARCADIS has studied the transport of dangerous goods for a 53 kilometre railway tunnel through the Alps between Italy and France. Transport of dangerous goods in both train and trucks on trains is foreseen in the future. Analysis, modelling and studies about the possible effects of an accident involving dangerous substances in the tunnel were carried out.

Both ADR and RID regulations enable the competent authorities responsible to restrict the transport of dangerous goods through tunnels. However, account should be taken of the tunnel characteristics, and a risk assessment including the availability and suitability of alternative routes and modes and traffic management should be conducted.

ARCADIS has studied the transport of dangerous goods for a 53 kilometre railway tunnel through the Alps between Italy and France. Transport of dangerous goods in both train and trucks on trains is foreseen in the future. Analysis, modelling and studies about the possible effects of an accident involving dangerous substances in the tunnel were carried out.

Analysing the dangerous goods

It is impossible to analyse the possible effects of an accident with all substances covered by the ADR/RID. The grouping method as proposed by the UN working group on the transport of dangerous goods (WP15) can be used to choose representative goods. For each of these goods it has to be determined whether the possibility of an accident in relation to the effects is acceptable and if safety measures can be taken to ensure that the required safety level, as determined by competent authorities is reached.

In a nutshell the following cases have to be executed:

Group A:

Dangerous goods in this group are prohibited for transport, so there is no need to study them.

Group B:

Dangerous goods that can provoke a big explosion. LPG is the product most transported good within this group. At least two types of incidents have to be considered: a leak of limited size leading to a vapour cloud explosion after ignition and a catastrophic failure of a LPG vessel leading to a BLEVE (Boiling Liquid Expanding Vapour Explosion). A combination of a leak and ventilation flow may result in an explosive cloud within the tunnel. For clouds of 10, 20, 50 and 100 metres in length (it is never certain when a cloud may be ignited), the blast effects must be modelled. A vapour cloud explosion can generate a blast that can cause victims up to a certain distance in the tunnel (for Lyon-Turin this distance was 4,200 metres). The blast of the simulated BLEVE can cause victims up to a certain distance on both side of the incident.

Group C:

Dangerous goods in group C are goods that can provoke an explosion (but not as big as group B substances) or a toxic leak. The dispersion of ammonia and chloride (both toxic gases) in the tunnel are representative to analyse the effects. Depending on concentration and exposure time the mortality rate for toxic cloud at a certain distance from the incident can be calculated. Toxic vapour clouds from toxic liquids also have to be analysed. Since evaporation from a liquid pool is a lot slower than evaporation of a pressurized gas, effects will not carry as far as the effects of pressurized gases.

Group D:

Dangerous goods in groups D are those that can provoke a major fire. To analyse possible effects of incident with goods in this group, the effects of a major fires (100 or 200 MW) in the tunnel must be modelled. Effects on passengers related to the temperature, the heat radiation and toxic concentrations of smoke can lead to additional measures in the tunnel (the ventilation principle).

Group E:

Since the products in group E are either less dangerous or transported in smaller quantities this group can be analysed on a lower detail level. Measures taken to reduce risks for the other groups are supposed to be efficient enough to take away the risks of this group.

Conclusion

Depending on the results of the analysis and the required safety level, additional safety measures might be required. This can be safety measures as:

  • Separating dangerous goods and passengers (by a minimal distance).
  • A ventilation system designed to reduce the toxic effect of gases by influencing the velocity and the direction of a toxic cloud.
  • An evacuation system designed to limit the surface of a possible pool to minimise evaporation and reduce the effect of toxic and flammable liquids.
  • An evacuation system designed to separate spilled dangerous goods and water (they can provoke a chemical reaction in some cases).
  • Special systems to protect passengers in the tunnel. For example, the ability to close the ventilation systems of trains passing through the tunnel, to ensure the safety of train passengers in case of a toxic leak.

For each safety measure its contribution to the objective can be determined, being: transport of dangerous goods with limited risks.

To contact the author, please email: [email protected]