Really efficient. They use less than 1/3 the energy of a freight train, per tonne per kilometre*, and in some places (like BC) the power mix is vastly greener for the pipeline than the train.
First, excellent story of the day, a Belgian brewery wants to put in a beer pipeline. This is awesome.
Second, my good friend Michael made a quip about this, saying "not all pipelines add greenhouse gases."
Well...the comment made me do some math.
1150 km is a funny distance to move something, but it happens to be the distance of the Trans Mountain (EDM - YVR) pipeline. Northern Gateway is actually more like 2000 km. For our purposes, longer transit distances make transport efficiency more important, but I did all the calculations based on 1150 km.
Showing your work is boring, so trust me:
It takes 18.3 kWh to move 1000 kg of light crude 1150 km in a pipe, courtesy this report on pipeline pumps and some math around the density of crude oil.
It takes 68.6 kWh to move 1000 kg of anything 1150 km in a freight train. Again, after mathing out the energy in a litre of diesel and some other conversions.
(1000 kg of fuel oil or diesel has an energy capacity of 13440 kWh. Again, I've done some unit conversion to get there.)
And it matters a bit that if run a pipeline through BC, it will be about 70% water-powered, thanks to our electrical grid power sources. Freight trains in BC are not water-powered.
So the question from a GHG point of view is whether the nonexistence of a pipeline causes about 0.5% less oil to be consumed (NOT transported to Vancouver; one fate of the "excess" oil is being sold off cheap into the local market: the price for crude oil in Canada is lower than it is in China, because we have an effective surplus of the stuff, no transportation needed.)** So unless a pipeline of this length causes oil extraction to rise a fair bit***, it probably earns GHG credits.
There's a lot of hand-waving here: I really don't know how responsive extraction is to price fluctuation, and the price difference is big: experts seem to think Alberta oil sells for about 20% less than it could if there was easier access to "tidewater" ports (that is, Northern Gateway and Keystone XL). On the other hand, that premium is large enough that even rail looks pretty viable. If Alberta isn't already shipping large amounts of oil by rail, it will once these pipelines die. What I'm saying is that we must not be blind to economics: lower the price of something, raise the demand. I don't know how much a pipeline will affect the demand for oil (though others have spent considerable time modeling this).
What I can tell you pretty conclusively is that if you intend to ship a quantity of oil through BC, you will burn less than 1/10 of the fossil fuels doing it with a pipeline than with a train.****
*Yes I am well aware I could express this dimensionlessly ("per weight per distance") but I think that's more confusing for a lay audience. Maybe I should have said "per gram per parsec" for lulz.
**Because the difference in green-ness is that moving fuel oil by train takes about 0.51% of the load, in fuel oil, to power the train. The pipeline takes about 0.03% of the load in fossil fuels (coal or natural gas in this part of the world) to move it, with the balance provided by hydro. So the wash is about 0.05%, call it 0.048% if you trust my figures.
***If my guess is right, the break-even is creating an increase in Alberta oil production equal to at least 0.05% of the amount expected to be shipped by rail if the pipeline is cancelled. Why is left as an exercise for the reader. Also, plenty of fudging about the fact that the Northern Gateway pipeline is 2000 km; If the alternative is moving oil by rail 2000 km to Kitimat, my figure would be way too conservative about the GHG break-even of the pipeline; it's more like 0.1%. That doesn't sound like much, but whether it happens is dependent on a bunch of factors. If, for example, Alberta oil production is at or near effective capacity, the addition of a pipeline might only increase the price the oil gets, not the production rate. On the other hand, the pipeline may well increase production. I don't know what the current pressures on oil sands production are. For all I know, it may currently hinge on how many Nova Scotians can be recruited to work at the Pizza Hut in Fort MacMurray.
****Don't misread this too badly: trains are amazingly efficient in terms of transportation efficiency. But pipelines are even more efficient.
PS: while researching this article, I discovered the Swiss once used electrically-heated steam locomotives, which is the most hilarious industrial hack I have ever heard of.
PPS: yes I need to install a footnote plugin for the site, I suck.