On Tuesday November 29^th the Trudeau government approved the Trans Mountain Expansion Project (TMX). I am someone who has researched the topic extensively and have a very nuanced  view on the project. I also have a strong desire that any discussion about the project be evidence-based. As such I tend to keep my ears open for erroneous information and in the last 24 hours I have encountered a number of commonly-held myths that are being spread by anti-pipeline activists. Given my desire to see a pragmatic discussion of the topic I have decided it was time to present a quick post debunking some of the most common myths presented about the Trans-Mountain Pipeline. Without further ado let’s get started on the debunking.

Myth: A Spill in Vancouver Harbour would pose catastrophic human health risk  

Now we all agree that a spill in the Inner Harbour of Vancouver would represent an ecological catastrophe. However, the activists are not satisfied with scare-mongering solely on the ecological sphere and have repeatedly talked about it as being a potential human health catastrophe as well. The basis for this myth is the City of Vancouver (CoV) May 27th Trans Mountain Expansion Proposal Summary of Evidence which, as I wrote about in two posts Questions about the City of Vancouver May 27th Trans-Mountain Expansion Proposal Summary of Evidence and More on that “Toxic Benzene Plume”.

In those two posts I demonstrate that the modelling exercise presented in the CoV May 27th Summary of Evidence represents extremely questionable science. In the second post I described the modeling as both “troubling” and “an outlier”. For those of you not familiar with science-speak that is not something upon which you want to base your evidence-based policy.

To explain in layman’s terms the modelers in the CoV May 27th Summary of Evidence used a sort of bait and switch through the use of a “pseudo-surrogate”. As I explained in my post:

As an analogy, imagine you were tasked with compiling a survey of the animal population of Vancouver. To simplify the survey you didn’t ask your surveyors to try to identify the dogs by species instead asking them to group the dogs by size. For a subsequent risk analysis you then assigned the pit bull as a “surrogate” to describe the behaviour of all dogs smaller than 2 feet tall identified in your survey. Would you then feel comfortable with the outcome of that risk analysis knowing that the analysis treated every Chihuahua it counted as if it were a pit bull for risk purposes? If someone subsequently warned you to stay off the street for fear of being attacked by “surrogate pit bulls”, based on this analysis, would you stay off the street? Well that is what they did in this report with benzene.

I concluded my post with this:

In the academic community there is a simple rule: if a new study runs contrary to a body of research then it is incumbent on the authors of the study to explain the discrepancy. Sometimes the new study is a paradigm changer, but most of the time it represents an outlier of dubious use in decision-making. Unfortunately, the Levelton report does not explain why its results differ so dramatically from the scientific consensus. More troublingly, it does not even acknowledge the existence of the body of research out there, including an almost identical modelling study, that came to such startlingly differing conclusions.

In the programming and modelling world there is an expression “garbage in, garbage out”. In the case of the CoV modelling the information they used as an input was not even close to appropriate for the analysis and as such the conclusion of the study is simply a fairy tale not even worthy of consideration. A bitumen spill in the Inner Harbour would be an ecological catastrophe, but would have very limited human health repercussions and certainly nothing like the repercussions suggested by the City of Vancouver and reported by the anti-pipeline activists.

Myth: Multiple Oil “Spills” on the Existing Trans Mountain Pipeline.

I have heard repeatedly in the last 24 hours that the Trans Mountain pipeline has had numerous spills and is therefore not safe. This myth takes advantage of the rather unusual definition of a “spill” used by the National Energy Board (NEB). To explain, under the NEB definition, a release that occurs into a spill containment facility is still reported as a “spill”. So imagine you had a wonky faucet in your bathroom that dripped into your sealed bathtub. Were the faucet oil that could potentially be considered a “spill” under the NEB definition. An examination of the data shows that most of those “spills” (almost 70%) involved releases to containment facilities. The point of a “spill containment facility” is to contain releases before they hit the environment and as such these “spills” posed essentially zero risk to human health or the environment. They represented limited releases into systems specifically designed to contain such releases.

As Kinder Morgan points out at their web site, in the last 35 years there have been three reportable spills from the Trans Mountain pipeline which escaped containment (were released into the environment (one the three being the responsibility of a third party the 2007 Westridge spill in Burnaby). Now let’s also remember that absent the TMX, future oil expansion will be along the rails. Now let’s compare those three spills to the three latest Canadian oil-transportation rail spills: Gogama, Galenas and Lac Megantic and ask ourselves how many deaths did the three Trans Mountain spills cause and how many deaths have oil-by-rail spills caused?  There are NEB “spills” and there are spills and the two should not be confused. When an activists suggests that a spill occurred the first question they should be asked is : how much of the product escaped to the natural environment. If the answer is none, then that is not what you or I call a spill.

Myth: Increased Risks to the Fraser River

Last night while watching Global News I heard a native leader saying he was going to fight TMX because of the risk it poses to the Fraser River. At that point I wanted to scream into my television set that he was looking at the the problem exactly backwards. As I have detailed in my blog the TMX poses a substantially lower risk to the Fraser River than the alternative (oil-by-rail). The reason for this is that our national railway system was built before our national highway system and as a consequence the rail lines that will carry those oil-by-rail trains mostly run along the sides of rivers like the Fraser. The Trans Mountain, which was built after the Trans-Canada Highway, runs well-separated from the river for most of its route. A derailment of a rail-train has a much higher probability of ending up in the Fraser than a spill from the Trans Mountain and when you do the numbers you come to realize that oil-by-rail is much more dangerous to the Fraser River than the TMX would be.

Additionally, remember that if the TMX is not built the Puget Sound refineries will still need crude oil and that crude oil will come by rail. That rail route runs along the headwaters of the Kootenay River and along the Columbia River. A spill on that route risks polluting those two shared river ecosystems as well. Think I’m overstating the risk? Well look what a happened adjacent to the Columbia River last June. It was only freak luck that the spill was caught before it hit the river. Think that was a freak event? According to the Guardian, including that incident, at least 26 oil trains have been involved in major fires or derailments during the past decade in the US and Canada. Finally if the TMX is not approved that rail traffic is going to increase substantially. The Puget Sound needs its oil and if it doesn’t come via pipeline, it will come via rail.

Myth: Substantially Increased Risks to Orcas

The latest argument against the Trans Mountain has been its purported added risk to the resident Orca population. Originally the argument went: increasing the number of tankers would increase the number of collisions with marine mammals and this could result in the extirpation of the resident Orca community. That trope was quickly demonstrated to simply represent a mis-reading of a single scientific article by someone apparently unaware of BC geography. The paper indicates that the Orcas are at high risk of collision in Johnstone Strait, which would be a problem if tankers were heading in that direction, which they are not. As for the increase in tanker traffic, the TMX tankers would represent an increase of 720 more ship movements in a Strait that sees 23,000 ship movements a year. This at a port that is engaged in a build-out that will expand ship traffic significantly. If acoustics are really a concern for the activists then rather than fighting the TMX, they should be protesting the Port of Vancouver’s expansion plans.

Never a group to let a good idea go to waste, once the collision myth was busted the activists turned to the risks to the resident Orca population posed by an oil spill. This myth comes courtesy of the Raincoast Conservation Foundation who prepared a study “Report on Population Viability Analysis model investigations of threats to the Southern Resident Killer Whale population from Trans Mountain Expansion Project. The study represents a modelling exercise to examine the effect of the TMX on the resident Orcas. Now, much like the City of Vancouver modelling, I was a bit surprised by the conclusions of the analysis and so looked a bit deeper into the methodology and like the Vancouver Study I discovered another case of garbage in, garbage out. The model itself seems sound and the statistical methodology was excellent. The problem with the exercise was the data used to generate the results. In this case the authors chose to use a very interesting source for their spill occurrence frequency input value. Specifically they relied on a “report” called Foschi (2014). Except when I go to the references I see that Foschi 2014 is not a peer reviewed journal article but rather a blog post by an interested engineer.

The blog post that serves as the critical data input for this modelling exercise uses values from the Trans Mountain TERMPOL 3.15 General Risk Analysis and Intended Methods of Reducing Risk (caution large file) to generate a likelihood of an incident. But the modelling exercise appears to use the wrong numbers from that blog post. In Foschi he relies on Table 34 (p83 of 454 on the pdf) to get his risk of oil spills. Now looking at the table the TERMPOL 3.15 authors presented several scenarios, a current (Case 0), a Case 1 (expansion with no mitigation) and then Case 1a and Case 1b (expansion with specific mitigations to reduce risk). Now the intention of the TEMPOL 3.15 report is to provide a description of the appropriate and necessary mitigation efforts associated with the increase in tanker traffic that would come with TMX. The conclusion of the report was that Trans Mountain make use of those mitigations. Those mitigations (including tugs etc….) were subsequently written into the TMX proposal as a requirement for the project.

Given that the mitigation plans were incorporated into the TMX proposal would anyone care to guess which of Foschi’s numbers were used for this modelling exercise? Yes you guessed it, the “no mitigations applied” number (Case 1) for spills. They used the very numbers that the authors of TEMPOL 3.15 suggest are not relevant (since the intention of the work is to describe the mitigations). But that is not all. Because “the reports do not clarify the average volumes or the distributions of those small spills” the authors tested the projected effects of frequencies of larger (>16,500 m3)  and smaller (>8250 m3) spills that are double those stated above” [highlight mine]. So not only did they use the “no mitigation applied” numbers but they then doubled the likelihood of a spill from the “no mitigations applied” scenario. Let me say that again to make it clear, they took a value that was recognized as being 4-5 times too high and then they doubled the frequency of those spills for conservatism. They then took those inflated numbers and inserted them into their model.

This is essentially like looking at the difference between a “no seatbelt” and “mandatory seatbelt” scenario in car crashes. If someone argued that we should ban cars because accidents for passengers without seatbelts is too risky your response would not be to ban cars. Your response would be to point out that seatbelts are mandatory and that modern cars have air bags as well thus any analysis that ignores the existence of seatbelts and air bags is not terribly relevant. Well sadly for the Raincoast Conservation Foundation, their analysis metaphorically argues that people are driving around without access to seatbelts or air bags and therefore driving is too dangerous. The model, which relies on the inflated risk of incidents, thus presents a similarly inflated risk of extirpation of the resident orca population.

Sadly, no one with the expertise to catch this bait and switch had read the report prior to it going to the TMX Ministerial Panel. As a consequence the flawed inputs were used to generate necessarily flawed outputs and those flawed outputs from that modelling made it into the TMX Ministerial Panel final report. So I suppose the opponents of the TMX Ministerial Panel were correct, maybe we should have been allowed to cross-examine presentations because then someone might have had the opportunity to debunk this flawed modelling exercise before it was broadcast to the world and treated as correct.

Looking at this short list of myths it is easily understandable why so many people are fighting the Trans Mountain pipeline. Given the amount of mis-information that is being spread over multiple media platforms it is simply impossible for the common Vancouverite to get a grip on the real risks of the project. Like I’ve said in this post garbage in, garbage out. Public sentiment is being swayed because the truth is being flooded out by the myths.

To conclude this post, I want to be clear, I am not saying that the project must go forward. As I note in my previous post, I have serious reservations about the project, However, I believe we need a fair debate on the topic and fair debates must rely on demonstrably sound evidence and not the scare-mongering that I have heard blasted across my social media feeds.

Author’s note: 

I have updated the text to better explain how the Foschi data was used in the modelling as my initial version was apparently a bit confusing.