|Bitumen seep at Primrose. Picture stolen from the Edmonton Journal.|
Since at least May, CNRL's Primrose Cyclic Steam Stimulation (CSS, or "Huff and Puff") project has been leaking bitumen to the surface near Cold Lake and Bonnyville. In total an estimated 6,600 barrels have leaked. The company has stopped steaming that part of the reservoir (under orders from the Alberta Energy Regulator, or AER) and the leak rate is now reportedly under 20 barrels per day and declining daily. It could continue leaking (at lower and lower rates) for months or even years. At the moment Primrose is CNRL's only producing in-situ oil sands project, although a number of others are in various stages of development.
Unlike the Total's Joslyn incident, where a SAGD well spectacularly exploded in 2006, this is a slow leak. Where the leak is coming from seems to be a pretty contentious issue. CNRL says they think it's a wellbore integrity issue, meaning it's simply a matter of drilling new wells or repairing the old ones - the bitumen can still be safely recovered. Others have suggested they may have fractured their cap rock, which would be much harder to fix, if it can be fixed at all.
I haven't seen the evidence CNRL is presenting to support its claim that it's a wellbore issue, but sitting here in my ignorance I'm leaning towards thinking it's not. Intuitively I would think that a wellbore leak would respond to shutting in more quickly than a broken top seal, and the response so far doesn't seem to be very fast. If this article is correct, the steam injection depth at Primrose is about 500 m, much deeper than was the case at Joslyn, and might explain why a similar problem (cap rock failure) could lead to a different result (slow leak versus catastrophic blowout).
What are the implications for CNRL? Well the company is apparently claiming the property will still produce an average of over 100,000 barrels per day this year but I think that's a stretch - their Q1 financials show they produced an average of 109,000 barrels per day from the project in Q1, and I don't see how they could stay over 100,000 without steaming for half a year, but maybe there's something I don't understand. If they're unable to stop the leak I think it's unlikely the AER will let them resume steaming and in the worst case scenario I suppose the entire project could be un-salvageable. That would be a significant setback, although probably not fatal. On top of the Primrose production, in Q1 of this year they (in a remarkable and confusing coincidence) also produced an average of 109,000 barrels of bitumen per day from their Horizon mine and 236,500 barrels per day of conventional crude elsewhere in North America (I think that's the number, it's confusing because they don't appear to split out in-situ bitumen from conventional and heavy North American production) and a further 35,000 barrels per day elsewhere in the world, namely the North Sea and offshore West Africa (all those rates are before Royalties). They also have significant natural gas production (1,150 MMcf/d, almost all of it in North America), bringing their total oil equivalent production to 680,844 boe/day before royalties, or 612,062 after. Losing 100,000 barrels per day would obviously not be fantastic, but more troubling perhaps is that bitumen production is their only consistently growing sector, so if Primrose's problems are intractable and the new developments can't pick up the slack they could be in long term trouble. I also imagine they're liable for a bunch of fines from the province.
What are the implications for the wider industry? Well if it is a wellbore issue, the message is clearly to have better completions on wells. If it's a cap-rock issue, it may mean companies need to look at reducing the pressures they're operating at, even in relatively deep horizons. This could significantly slow down production rates. If Primrose's problem turn out to be related to the use of CSS (rather than SAGD), it could further cement SAGD as the thermal stimulation method of choice in the province.