Constable Kerry Schmidt, speaking on behalf of the OPP investigation into the fiery tanker truck collision of yesterday, reported that investigators now believe that both of the vehicles that were engulfed by the fire on Highway 407 on the outskirts of Toronto were travelling westbound and therefore both were projected over the concrete median barrier into the eastbound lanes. While the OPP have all the evidence that neither I or anyone else can examine, it should be noted that such a happening would be extremely rare. If the median barrier was 110 centimetres tall it would be just slightly taller than the roof of a typical passenger car that might be in the range of 140 centimetres above the ground. An SUV might be in the range of 160 centimetres and pick-up trucks with high suspensions would be even taller. More importantly the vertical centre-of-gravity of any of these vehicles would be substantially less than the height of the median and this is why it is very difficult to project a typical, light vehicle over the concrete median. It was speculated that somehow the two vehicles became entangled and the tanker truck could have pulled the smaller vehicle over the barrier with it. Fact can sometimes be stranger than fiction but one must recognize that such an occurrence would be very rare and I have not seen such an occurrence in all my 38 years of studying thousands of major collisions. The technical term used in computerized simulation software is called “snagging” and there are options to invoke such conditions when running a computer simulation. However, invariably, such snagging occurs in horizontal motions when, for example, a stiff wheel becomes jammed inside the crushed wheelwell of the other collision partner. The difficulty with dragging both vehicles over the barrier is in producing enough of a snag, and enough of a vertical force, to get the smaller vehicle’s centre-of-gravity over the wall.

There has to be some obvious reason why the OPP have come to their conclusion. This may be from witness information or from specific physical evidence they may have found on the near side of the barrier, for example, showing a paint scraping or some other evidence that could be matched to that smaller vehicle. Because both vehicles were likely fully consumed by the fire it would be difficult to obtain such detailed evidence from the vehicle bodies themselves. Without having an opportunity to examine the evidence I can only commit to my comment that propelling both vehicles over this relatively tall barrier is difficult to imagine.

Further study of the dashcam video using a video editing program suggests that, from the instant the truck began to pitch over the barrier up to the time that the first evidence of visible flames was only about 3 seconds. It is difficult to be precise here because the view from the dashcam is from a substantial distance and other westbound vehicles were partially blocking the view. However, this 3 seconds isĀ  a very short time. It is more understandable if a leakage developed some time after the vehicles came to rest and something ignited the fumes. The fact that this fire erupted so quickly is more of a concern because it could demonstrate a vulnerability of fuel tankers that may not have been known up to this time. It needs to be identified what parts of the truck were found on the near side of the barrier to establish what parts became detached before the pitching over the barrier and whether those parts can explain the early fractions of a second when some separations might have occurred leading to the opening of the fuel into the environment. Again these are matters that experienced investigators would pay close attention to.

Another matter of importance is the study of the initial incident that led to the truck’s dramatic motion across the westbound lanes and toward the barrier. From studying hundreds of loss-of-control collisions I have observed that the initiation of visible tire marks that indicate a loss-of-control occur substantially after the event that led to that loss-of-control. In other words, whatever police find in terms of tires, those marks are not likely to identify the location where the emergency condition occurred. There are exceptions where a very sudden failure of a mechanical component produces an immediate gouge, scrape or tire mark. Much depends on the experience of police investigators in their ability to understand the physical evidence they are dealing with. One cannot simply walk into the role of accident reconstruction by taking a few weeks of courses without having a solid number of years of experience in examining and interpreting physical evidence. Those police who are involved in Collision Reconstruction units are introduced to a lot of physical evidence in a very short space of time and therefore they can become adept at interpreting such evidence in 5 or 10 years time. Unfortunately such officers become re-deployed to other departments just as they become the most useful in their field. But that is another matter that needs discussion on another day.

A number of vehicles were in the vicinity of the far right lanes where the truck was located just before it made its dramatic left motion. This may provide the opportunity for police to piece together the witness accounts into something that could be useful. Again, from my experience of conducting many such analyses, witness information can sometimes be good and accurate while in many cases it can be quite scattered, imprecise and worse. The wisdom of a dedicated and detail-oriented analyst can separate what reported evidence is reliable.

It should not need to be emphasized that the loss of innocent lives requires that we do our utmost to prevent such tragedies in the future, where possible.