On pipelines, oil-by-rail, and the relative risk of catastrophic spills in the aquatic environment

As my regular readers know besides this blog I also write a blog at the Huffington Post. It typically consists of shorter versions on my pieces here. Well my most recent post on the Trans-Mountain Expansion project has received a lot of feedback, particularly since the post went online shortly before the Husky Oil Spill. As such, I wanted to take a few moments to address some points that I have made repeatedly in the comments at the other blog but have not had the space to flesh out properly there. Specifically, I do not want to talk about the relative absolute risk of pipeline spills but rather the relative human and ecological risk of spills from pipelines versus oil-by-rail. To explain my intention, I am going to use the tools of risk assessment to help you understand why I feel strongly about getting oil off the rails and into pipelines.

As I noted above, the Husky Oil Spill is headlining the news. As reported, the spill occurred proximate to the North Saskatchewan River with about 200,000 L – 250,000 L of heavy oil being released. While this seems like a large volume remember that we are talking in relative terms here. A unit train for transportation of oil-by-rail consists typically of between 70-120 rail cars with each rail car carrying approximately 600 barrels of crude oil (approximately 95,000 L). So the Husky spill consisted of a little over two rail cars worth of product. The North Saskatchewan River has an average flow of about 238,000 L/second so the entire spill consisted of about a second’s worth of flow on the River. Had the material spilled been diesel or gasoline, the resultant spill would have dissipated to non-detectable concentrations downstream; but this was not a gasoline/diesel spill. Instead it sounds like the spill was heavy oil (possibly dilbit) and diluents (likely condensate). As I have written previously, the resultant volume would have an initial density around 0.92 – 0.94 mg/L but as the lighter materials separated out the heavier oil would eventually reach a density approaching 1 (the specific gravity of pure fresh water) and would have the potential to sink. I go into a lot more detail about what happens next at my earlier post. On a positive front, since the material is heavy hydrocarbons the material is very hydrophobic and not very soluble so much of the material should be recoverable using the right technologies. Suffice it to say this is a bad spill and will take a reasonable amount of work, time and money to clean. In the process some valuable habitat will be damaged and a lot of downstream communities will have some serious worries possibly for some time to come.

 The thing that is unusual about this pipeline spill was its proximity to water. You see that is one of the big benefits to pipelines: their routing. As every person who went through the Canadian school system can tell you, the construction of the CN and CP rail lines represented marvels of their era. These two railways cross some of the most challenging terrain in North America. To do so in that era, these rail lines took advantage of natural terrain features which when going through mountains means rivers. Just take a look at the CN rail map (caution large file) or the CP rail map. Both run the majority of their length immediately adjacent to rivers. The Trans Mountain pipeline, however, was built over a century later and was built after the Trans-Canada Highway and many of the rural highways had already been constructed. As a consequence, it runs far less of its length along river edges. Consider the pipeline in communities like the lower mainland. The current pipeline route pretty-much avoids the rivers and runs straight down the valley; the rail lines, meanwhile, run pretty-much right along the river for much of the length of the Valley. So what does this means for safety? Well this is where I bring in my background in risk assessment into play.

As I have written previously, in a risk assessment for there to be an unacceptable risk to human or ecological health the following conditions need to exist:

  • chemical or compounds must be present at hazardous concentrations;
  • human or ecological receptors must exist; and
  • an exposure pathways must exist that allows the chemical compound of interest to interact with the receptor resulting in a dose that poses an unacceptable risk.

Now for a relative risk assessment between oil-by-rail and pipelines we can all agree that for both modes of transport the compounds are present at hazardous concentrations and human and ecological receptors exist. The critical difference between the two is the “exposure assessment”. In this case we ask the question: how much of the route poses a direct risk to rivers and aquatic ecosystems? For the pipeline we would need to consider  every instance where the pipeline crosses next to, or under, a river or a stream. For rail, which runs along the side of the river, we have a pretty much consistent risk of 1 (i.e. if the train has an accident along a riverbank the chance of it affecting the river in some way is 1). In a risk assessment we then use this information to calculate a relative risk.

Now we already know that oil-by-rail has a 4.5 times greater risk of incident so let’s now do a back-of-the envelope calculation of the increased risk to aquatic ecosystems associated with oil-by-rail versus pipelines. Let’s take a wild guess that the pipeline in the Fraser Valley has 5% of its length crossing under or adjacent to a river/stream (that number an educated guess and may be easily off by an order of magnitude). Then the increased relative risk for the oil-by rail to aquatic ecosystems would by 4.5 x divided by the difference in exposure (here it would be 0.05/1). The resultant relative risk would be a factor of 90 times. So from a relative risk perspective the oil-by-rail is not 4.5 times greater of an incident affecting aquatic ecosystems but rather 90 times greater. Once again I am using estimates to help you understand where I am coming from. What this simple example shows is that from an aquatic perspective the risk of serious impacts by rail is not simply 4.5 times higher but could be in the range of 90 times higher. I’m sure someone more familiar with the routing can provide better numbers but I think you can see that by using the tools of risk assessment we can get a better understanding of the relative risk of these two modes of transportation.

Another consideration not included in the risk calculus involves the storage facilities en route. As I pointed out in a previous post, the various pump facilities and storage yards along the Trans Mountain route are dedicated facilities, each of which includes dedicated spill containment facilities. These facilities are typically bermed and lined with materials that will not allow spilled oil to escape. As discussed 69.5% of the “spills” during the operation of the pipeline have been in facilities with spill containment and thus did not pose a serious risk of escapement. The rail system doesn’t use similar containment facilities along their rail lines. When an oil train is pulled up along a siding or in a storage yard they can’t install a bermed facility with liners to catch spills. Rather the trains can  sit unattended or only attended simply by railway staff who have other responsibilities as well. That is what happened at Lac Megantic.

I want to keep this post relatively short so will stop here. To conclude I want to reiterate a few things. First, I want to make it clear that I am not trying to play down the Husky Oil Spill. I don’t know enough about the current situation of that spill to give a reliable estimation, but I can assure you that the clean-up will not be over next week and if a lot of the material has become affixed to river sediments and sunk to the bottom, then the clean-up is going to take time and be expensive. That being said, this spill represents the equivalent of two tanker cars worth of material. Imagine if it had been a unit train that had left the tracks adjacent to the river? The volumes would have been 50 times greater. Moreover, as I point out for the province of BC, while the risk of incident is 4.5 times higher for transportation via rail over pipeline; the relative risk is even higher for rail, because the rail system is not dedicated solely to the transportation of fuels and because the rail lines are situated far closer, for a longer proportion of their length, to the more fragile aquatic habitat. If you are a community that depends on the health of our rivers, you will want to get as much of that oil off the rails and into pipelines as soon as possible. The numbers make that clear.

This entry was posted in Canadian Politics, Energy East, Oil Sands, Pipelines, Risk, Risk Assessment Methodologies, Risk Communication, Trans Mountain. Bookmark the permalink.

16 Responses to On pipelines, oil-by-rail, and the relative risk of catastrophic spills in the aquatic environment

  1. Douglas MacKenzie says:

    The Husky spill is certainly an unfortunate incident which will have an effect on wildlife and drinking water for downstream communities for 3 weeks or so. Husky will pay for the cleanup and face large fines to set an example to the industry and as a display of regulatory governance. But practically speaking the fear of oil pipeline spills is exaggerated to the point of irrationality. No human life is lost in an oil pipeline spill. For the average pipeline spills in a year, more frogs are killed in Biology classes, more mammals are roadkill in 100 km. of any highway in Canada, orders of magnitude more birds are killed supplying bar patrons with hot wings. There is more heavy oil in the asphalt pavement of the parking lot of the average box store than in the average oil spill…but these don’t have the fear factor that make good news stories….

    Liked by 1 person

  2. Pingback: The Husky oil spill, its effects on the pipeline debate and a thought experiment about a world without fossil fuels | A Chemist in Langley

  3. Pingback: On the absence of intellectual rigour and honesty in the pipeline debate | A Chemist in Langley

  4. Pingback: Debunking some activist myths about the Trans Mountain Expansion Project | A Chemist in Langley

  5. Pingback: On Site C, Pipelines and Noble Cause Environmentalists | A Chemist in Langley

  6. Pingback: What are the real marine risks of the Trans Mountain Pipeline Expansion? | A Chemist in Langley

  7. Pingback: Trans Mountain serves Consumers | Friends of Science Calgary

  8. Pingback: Debunking another compilation of Trans Mountain pipeline myths | A Chemist in Langley

  9. Pingback: Let’s be clear, the actions of the self-styled “water protectors” will place the Pacific Northwest at greater risk of a major oil spill | A Chemist in Langley

  10. Pingback: On the environmental and social consequences if the Trans Mountain pipeline expansion project fails | A Chemist in Langley

  11. Pingback: Environmental Absolutists are going BANANAs and it will hurt, not help, the environment | A Chemist in Langley

  12. Pingback: Another look at the Trans Mountain Pipeline Expansion Project from the lens of a pragmatic environmentalist | A Chemist in Langley

  13. Pingback: Revisiting activist myths about the Trans Mountain Pipeline – or Why Climate leaders may sometimes need to build pipelines | A Chemist in Langley

  14. Pingback: Another day, another flawed CCPA report, this time about the Trans Mountain Expansion Project | A Chemist in Langley

  15. Pingback: Another Day, Another Flawed CCPA Report, This Time About the Trans Mountain Expansion Project – HoweStreet

  16. Pingback: Evaluating what the new Canadian Energy Regulator report actually says about the viability of the Trans Mountain Pipeline | A Chemist in Langley

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.