The Belgian University of Ghent has raised serious concerns about the reliability of a method used to determine the type and amount of fuel involved in the fiery explosion following the deadly Tempe train crash in February 2023.
In a written briefing sent to the Hellenic Air and Rail Accident Investigation Authority (HARSIA) back in January, the university highlighted a “high degree of uncertainty” in conclusions drawn using computational simulations.
The evaluation centered around a modeling approach employed by Greek researcher Kostas Lakafosis, who used computational fluid dynamics (CFD) and Fire Dynamics Simulator (FDS) software to analyze the massive fireball seen in video footage of the accident.
While the Ghent team acknowledged that FDS is generally appropriate for modeling fire-related scenarios, they stressed that missing critical data –like precise fuel loads, atmospheric conditions, and initial ignition details – makes the outcome of such simulations highly speculative.
The analysis noted that although certain elements, like the spread of the fireball and tank fires, fall within FDS’s validated range, the method can’t definitively identify the fuel type or quantity due to gaps in evidence.
The lack of a clear visual record and uncertainties about the commercial train’s cargo further complicate any reliable reverse-engineering efforts.
The University of Ghent concluded that numerous input scenarios could result in simulations that visually match the real explosion, but each with different underlying assumptions.
This significantly weakens the credibility of using this modeling alone to draw conclusions about the origins of the fireball.
Despite being available to Greek authorities since January 28, these findings raise ongoing concerns about the thoroughness of the investigation.