Firstly, may we, the Safety First Association and its members express our shock and sincere condolences to those who lost family members due to the gas tanker explosion on 24 December 2022 in Boksburg. Our thoughts and prayers are with you all and we hope and trust that those who are injured and traumatised will recover soon.

In South Africa, we have various vehicles transporting hazardous materials daily. Most people have no idea of the risks related to the types of dangerous and hazardous goods being transported on our roads, what could happen should an accident event occur, and even how they are transported. Hazardous gasses are transported in pressurised LPG tankers, carbon dioxide tankers, oxygen, chlorine and acetylene tankers. While others may be transported in a liquid form in pressurised tanks or cylinders on trucks and even bakkies. Then we have the large tankers carrying flammable liquids such as petrol, paraffin, solvents, diesel and jet fuels, etc. in unpressurised tanks. There are also trucks carrying solid and powdered materials like ammonium nitrate and explosives. Corrosive acids and alkalis also have to be transported. Any of these tankers or vehicles can and have been involved in unfortunate incidents in the past.

I waited a few days since the Christmas eve gas tanker explosion incident for incident information to be publicised, so that a logical incident assessment can be made, based on the information available and my insights as an experienced incident investigator. In time, the other formal incident investigations should provide the actual incident causes going forward.

What happened?

Stitching together the press and media reports let’s perform an assumed gas explosion incident scenario assessment.

The incident background involves a gas tanker reported to be travelling from Richards Bay to Botswana in a contracted haulage carrier company’s Hazchem (hazardous chemical risks labelled) designated tanker vehicle with non-company employee driver who was reported as a suitably trained, licensed and professional permitted driver with reported some 7 years of driving experience, apparently obtained from a third-party Employment Agency. At this stage we are unaware of the hazardous goods travel route and had the route been formally risk assessed and approved for use. Secondly, we are still uninformed to what extent the tanker driver was informed and instructed on the assessed and approved travel route, which should have at least one overnight rest stop due to the travel distance involved.

It is assumed the travel route was not along the N3 from KZN, but from Richards Bay via Mpumalanga, which is likely why the tanker was passing through the Boksburg area. I was recently in the East Rand area and got lost as the road signage is poor to non-existent. That being the case, the employment agency driver could have easily got lost and taken a wrong turning, ending with the tanker being caught under a low height capacity bridge with poor warning signposts. A pre-incident bridge photo that has been published shows a small and poorly visible warning sign displayed on the bridge parapet railing. The bridge’s concrete beam also shows damage where numerous previous vehicles would have impacted with it. Furthermore, the published photographs of the bridge site display a steep road profile decline entry into the bridge, which means the long tanker has its back wheels up the slope adding to the tanker’s height on entering under the bridge. Another published photograph of the tanker wedged under the bridge shows possible evidence of fire flames, so the fire ignition may have been caused as the tanker got stuck under the bridge. Apparently, the driver realising there was a gas leak and a serious incident risk immediately called for emergency help. He was also said to be involved in deploying barrier tape and warning the spectators to evacuate the area as gas was escaping. But without any heeding his warning, they became victims. Such actions suggest the driver was risk trained and so took the appropriate incident prevention actions.

On arrival of the emergency services (fire brigade and police), they automatically become the incident site custodians to manage the site and the risks. This would involve managing the gas leak and fighting the fire when it occurs, as well as managing and stopping road access to the site, and due to the gas risk enforcing the evacuation of the area of all the bystanders, etc. But, it appears that no enforced bystander or other evacuation was implemented.

Why did the gas tanker catch fire and explode? Safety things to note:

1. LPG is stored as a liquid under pressure in pressure tanks, tanker cylinders and in the domestic and other gas cylinders used by us all. When the gas pressure is released the LPG liquid rapidly vapourises into a gas that occupies about a 270 times larger gas volume than a 1 litre of LPG’s liquid volume. It is believed that the driver knowing the risk therefore called for the emergency services help to manage the incident.

Note: The Emergency Management Services adopt the commanding and managing responsibility of any incident site on their arrival.

2. LPG is heavier than air so it will sink and accumulate into the lower lying areas under the bridge, and when the gas vapour is at a 2% to 9.5% mixture in air it only needs a nearby spark, a spectator cigarette or something at some 450°C to ignite it.

LPG being a heavy gas would raise oxygen out from under the low bridge area, giving an oxygen deficient atmosphere such that nearby vehicles were not able to be started, as was reported by taxi driver/s who lost their vehicles in the explosion. And for this same reason, the driver was hospitalised as he was reported to suffer from an oxygen deficiency exposure.

3. On ignition the accumulated leaking gas, from an unknown source, possible from a hot exhaust, electric spark or cigarette, etc., the leaked LP Gas will flash back to the leaking tank, heating the tank (as the flames and heat, about 1 980°C, was confined against the tank against the bridge’s concrete deck base), where pressurised LP liquid in the tank is heated, and with reducing gas leak pressure, the LPG liquid rapidly changes to LP Gas which will over pressure the metal-weakened fire-heated tank until the tank ruptures causing a fire related BLEVE (Boiling Liquid Expanding Vapour Explosion). This then generates a massive high heat and energy fire ball of several hundred meters in diameter, due to the tanker volume of LPG involved.(Note, steel melts at about 1600°C).

Considering this assumed incident sequence of events, now leaves us with questions from which lessons can be learnt.

Incident lessons that can be learnt

Lesson one:

The authorities and the public are relatively unaware of the risks related to the transportation of hazardous materials through their town and community areas – in this tragic case Liquid Petroleum Gas (LPG). What are the dangerous and hazardous goods that are transported daily in your area? Are they being transported through your community high-risk areas? Have specific approved high-risk transportation routes been established and their use enforced? Are these vehicles suitably approved and permitted to transport such hazardous goods?

Lesson two:

Was there an adequate hazardous good transportation route and driver knowledge and capacity assessments done? Currently, we are unaware if there was any transportation route risk assessment performed by the transport contractor. Was the route selected a normal transportation route or was it changed due to a holiday season quiet route from Richards Bay to the Reef, so to avoid the busy N3 traffic routes via Durban? It seems possible that the route selected involved the driver travelling through Mpumalanga ending in Boksburg, where I recently got lost travelling from the N17 towards the N12 as the road route signage is poor. Question: was this route the normal tanker route or an adapted one so as not to subject the festive season traffic on the N3 to a hazardous load risk? Had the authorities perhaps placed a limitation on the heavy transport haulers using the N3 during this festive season period, and thereby forcing the transport contractor to use other less familiar and possible non-risk assessed routes? There are many lessons to be learnt within this second lesson learning point.

Lesson three:

Assuming the driver got lost, took a wrong turning and got his tanker wedged under a low level clearance height bridge with warnings poorly marked. Press photos show the tanker wedged under the bridge, with no obvious fire ignition showing, although it is obvious that some valve on the tanker top was damaged and so it was leaking LPGas into the atmosphere. The knee jerk response is/was to blame the driver for the deaths (now standing at 34). Yes, he wedged the tanker. No, he did not deliberately or intentionally start the fire. He apparently acted responsibly by highlighting the gas risk and so implementing “keep clear” precautions until the emergency services arrived. He was not responsible for the site after the emergency services (fire brigade and police) arrived and started managing the site. He had no authority to order a site evacuation. However, the general feeling is everything that happened is his fault. This lesson suggests there are more root causes to this incident than just putting the blame on the tanker driver.

Lesson four:

Was the fire-fighting adequate? LPG is an E class fire which is normally extinguished by shutting off the gas supply or by firefighting using dry powder, or carbon dioxide, or foam into the gas stream to extinguish the gas fire. In this incident, water hoses were used in an attempt to keep the tanker’s tank cool, but this would not extinguish the gas fire. Unfortunately, the gas leak fire was against the bridge base, so the flames and fire heat were in contact with the metal tank surface now being heated, thereby softening the metal until the tank under pressure ruptured, releasing the gas liquid to cause a BLEVE and its huge fire ball and explosion which destroyed lives and property. The lesson is that fire-fighting must always consider the material burning and the environment impacting on extinguishing a fire.

Due to the gas leak fire location in this scenario, there was a slim chance of extinguishing the fire whereas a tank rupture and explosion was a likely outcome. Therefore a full area evacuation was required but was not ordered and/or effectively implemented, resulting in damage and fatal injuries from the BLEVE, which caused an explosive shock wave and intense fire balls impacting everything in at least a 100m diameter area. The explosion shock blast caused the roof and ceilings of the nearby Tambo Memorial Hospital to be blown off with some walls collapsing. The bridge, railway lines and nearby vehicles were destroyed. Surrounding houses were damaged. People in close proximity were either burned or killed.

Lesson five:

Did the bridge display adequate and obvious warning signs indicating its low clearance and clearance height dimensions? Media pictures show a poorly visible bridge warning sign which many drivers had obviously not seen because the bridge’s concrete beam can be seen as damaged multiple times from vehicle impacts. That suggests the warning signage was totally inadequate and should have been upgraded to suitable visible signage by the municipal authorities. With the height warning signage being in a poor state, the driver could easily have misjudged the height clearance for his tanker truck, which was evident in the photo of the truck under the bridge. Knowing that the signage was inadequate, why did the local council (traffic department) not improve the low bridge warning signage with clear visible warnings on the approach road?

Will the bridge be rebuilt as per the destroyed one, another low-level clearance bridge? More than likely, because of the railway lines alignment issues. If so, more bridge vehicle collision contacts can be expected. An over-the-rails design bridge is required to eliminate this low bridge hazard, but at a significant cost.

Lesson six:

Are the local authorities and emergency services up to speed to handle such incidents? The media has been asking such questions for a while now, as it is known that suitably trained and experienced staff shortages and equipment supply and maintenance funding issues exist and are known concerns in nearly every South African municipality and town council. Was fire-fighting foam available for use at this incident site? We also ask, when will the destroyed fire water tanker be replaced following this incident?

Conclusion

This assumed incident risk assessment highlights that many factors, like evidence, forensics, witness and response statements, etc, will need to be considered, by the hopefully experienced incident investigators, during the formal incident investigation and its issued report.

Let us hope that some points from this incident risk assessment of assumed questions and lessons will be considered during the investigations and implemented going forward in memory of those who tragically died.

By Leighton Bennett, the Safety First Association Chairperson, and a professionally registered OHS Professional and Risk Management Consultant.

www.safety1st.co.za