Kinder Morgan’s “enhanced” spill response regime: A closer look

This post is part of an ongoing series in which MLA Andrew Weaver will be sharing key information from inside the National Energy Board hearings on Kinder Morgan’s Trans Mountain pipeline proposal. To see previous posts, please click here.

What would happen if there were to be an oil spill on our coast?

How much of the oil could we actually expect to recover?

Kinder Morgan tried to address that second question in its application. Unfortunately, I found their conclusion to be rather unbelievable.

As a part of their application, Kinder Morgan submitted a report done by EBA, A Tetra Tech Company. The report was designed to evaluate Kinder Morgan’s enhanced oil spill response regime by simulating an oil spill.

The model they used to run this simulation is proprietary, so we have no way of assessing how valid or accurate it is. But we do know enough to raise some serious questions about the results.

 The Oil Spill Simulation

When EBA ran its computer model, they started by outlining some basic assumptions. And that’s the key. In science, your projections are only as good as your initial assumptions. Unfortunately, the assumptions they chose are not representative.

Here’s what I mean:

When EBA ran its oil spill simulation, they had to pick a day of the year when the spill would occur. They also had to define what the weather was like that day, how quickly responders could get to the scene, when they would start their work, etc. All of these choices are called assumptions, which are converted to model input data. That is, using these assumptions, the model then has some hard data that it can use for its calculations.

Now, one thing we know in science is that the assumptions you choose can significantly influence the final result. For example, if you assume that responders can cleanup 100 tonnes of oil every hour then the final result will be much better than if you assume they can only cleanup 1 tonne every hour.

Questionable Assumptions

So what assumptions were used?

For starters, a pristine summer day in August was selected. They assumed that there would be 20 hours of sunlight to facilitate the spill response, that adverse weather conditions (such as waves and wind) would not prevent or complicate response in any way, and that there would be no toxic or explosive hazards preventing first responders from immediately approaching the spill. Finally, they assumed that it would take thirteen hours for all of the oil in the ship to be released into the surrounding environment.

Let’s unpack this.

First of all, we know from weather data provided by Trans Mountain (p. 366) that weather conditions along the route prevent spill response for 10% to 40% of the year. On top of that, we also know that even when spill response is occurring, the response becomes significantly less effective whenever waves are taller than 1m. Anyone who has been out on the water knows that we often have waves that are taller than 1m.

Secondly, to assume that it will take thirteen hours for all of the oil to spill out of two tanks seems like an arbitrary assumption. I asked Kinder Morgan to explain the reasoning behind it, but received no actual justification.

Finally, lets calculate how much sunlight there is in August. We know that the longest day in August will occur on August 1, as that is closest to the summer solstice (June 21). It’s relatively straightforward to plug the latitude and longitude of Victoria into the US Naval Observatory Astronomical Applications software to find that on August 1st, 2014 (for example) there was 15 hours and 2 minutes of daylight. Even if we stretch this to occur between the beginning and end of civil twilight, we only get 16 hours and 7 minutes (see figures below). This sure doesn’t seem like 20 hours to us. In fact even at the summer solstice, the longest day of the year, there is only 16 hours and 15 minutes of sunlight extending to 17 hours and 31 minutes if we include the beginning to end of civil twilight. In fact, if you wanted 20 hours of sunlight on any day in August you’d have to be at Tuktoyaktuk’s latitude (69.5° N). And that’s hardly relevant to conditions off southern Vancouver Island.

a)       b)

Figure: (a) Sunrise and sunset calculations for August 1, 2014 and (b) the summer solstice (June 21, 2014).

The Results

After selecting pristine conditions, the report concluded that 44.5% of oil could be recovered at sea outside of the immediate containment area and a further 18.6% could be recovered within the containment area thanks to Kinder Morgan’s enhanced spill response regime. That result may seem pretty good, although you might be wondering what happened to the other 36.9% of oil that was spilled.

But the unfortunate reality is that EBA’s result appears unbelievably high as far as oil spill recovery goes. According to the Federal Expert Tanker Safety Panel, on average only 5-15% of oil is actually ever recovered from a spill, even in optimal conditions.

So is Kinder Morgan’s plan just that good?

The short answer: Unlikely. Rather, by picking ideal assumptions, EBA came up with a result that isn’t necessarily representative of what would actually happen in a real spill.

Given this, I asked Kinder Morgan to redo their model analysis with more representative results. Unfortunately, they refused.

I can’t help but come back to the same basic point: British Columbians expect better.

And unless Kinder Morgan is going to step up and take oil spills seriously, they will never earn the social license they need to build their pipeline.