Thanks. Interesting article. I assume the water that gets pumped in is supersaturated with salt to minimise wear? I wonder if the draw down limit is, in part, a publicly available limit and the rest is earmarked for military, security and emergency services. The vagueness of the limit smells like operational security.
Are they actually lowering sonar devices into the caverns, to image the shapes? It seems like it should be possible with where drone tech is now -- you probably could even use a fiber lead to maintain contact with it while it's down there.
Based on the challenges and limitations of the storage mechanisms, it seems like drawing down the SPR should be considered only as a last resort during an emergency. Not something to be done on a whim based on political expediency.
The mechanics here explain something the policy argument usually skips. A single-cycle cavern is a one-time option, so the reserve's real size is not 714 million barrels, it is the number of times you can still sell into a crisis. Add the dismantled degasification facility and the barrels deliverable in a warm month are fewer than the barrels owned, which is a delivery risk a refiner prices into its bid rather than a headline inventory number. Below 300 million now, roughly 120 million released, and every drawdown makes the next one physically worse. That is a depleting asset, not a buffer.
With the advent of fracking the use of drilled but uncompleted wells (DUC) has become more common. Is this another way the SPR could be achieved for NG and light petroleum?
What surprised me is how closely this tracks the Federal Helium Reserve. There the public debate sounds strategic, but the real limit is reservoir pressure. Draw the storage formation too far down and the remaining gas cannot be produced fast enough to matter.
We stored the nation's emergency oil in a hole that shrinks when you use it. Then we spent 50 years arguing about the deductible. America's energy plan: a panic button with a weight-loss program.
What are the economics of the SPR? eg if creating and maintaining the SPR roughly doubles the cost of each barrel, and then we can only draw down half of the SPR, then we have already "pre-shocked" ourselves with a 4x price increase. How much does this make sense, vs (eg) just keeping a pile of money around to stabilize the price by guaranteeing a buyer of domestic supply? I don't love price supports but I also don't love big holes in the ground that eat half the oil we put in them
Good point that we are generally only withdrawing in high price regimes. Still, if we can only pull out half of the oil, the we are closing in on $1/month/barrel, so … that starts to add up. Thanks for inking up a napkin for me :)
A public service. Thank you.
Thanks. Interesting article. I assume the water that gets pumped in is supersaturated with salt to minimise wear? I wonder if the draw down limit is, in part, a publicly available limit and the rest is earmarked for military, security and emergency services. The vagueness of the limit smells like operational security.
Are they actually lowering sonar devices into the caverns, to image the shapes? It seems like it should be possible with where drone tech is now -- you probably could even use a fiber lead to maintain contact with it while it's down there.
This was really interesting. Thanks.
Based on the challenges and limitations of the storage mechanisms, it seems like drawing down the SPR should be considered only as a last resort during an emergency. Not something to be done on a whim based on political expediency.
Great piece; it answered a lot of my questions!
I thought China had huge storage facilities and had stopped importing crude oil while the SoH was closed. Am I wrong?
They do have a lot of petroleum storage, but as I understand it it's a lot more distributed.
Salt cavern dynamics and geological storage are wild engineering feats. People forget that national scale infrastructure relies on basic rock mechanics and fluid pressures. Wrote about simulation boundaries and modeling physical systems here: https://gzambrano.substack.com/p/the-physics-of-what-a-model-decides-to-forget-information-approximation-and-the-limits-of-simulation
The mechanics here explain something the policy argument usually skips. A single-cycle cavern is a one-time option, so the reserve's real size is not 714 million barrels, it is the number of times you can still sell into a crisis. Add the dismantled degasification facility and the barrels deliverable in a warm month are fewer than the barrels owned, which is a delivery risk a refiner prices into its bid rather than a headline inventory number. Below 300 million now, roughly 120 million released, and every drawdown makes the next one physically worse. That is a depleting asset, not a buffer.
With the advent of fracking the use of drilled but uncompleted wells (DUC) has become more common. Is this another way the SPR could be achieved for NG and light petroleum?
Interesting, I worked at a potash "mine" in Saskatchewan that worked this way, pumping hot water down and pumping out with potash that was then dried.
Natural gas is also stored in similar caverns in Saskatchewan though I don't know the history of them.
A reserve that only gets drained is a slogan. The barrels already left.
What surprised me is how closely this tracks the Federal Helium Reserve. There the public debate sounds strategic, but the real limit is reservoir pressure. Draw the storage formation too far down and the remaining gas cannot be produced fast enough to matter.
We stored the nation's emergency oil in a hole that shrinks when you use it. Then we spent 50 years arguing about the deductible. America's energy plan: a panic button with a weight-loss program.
So insanely fascinating
What are the economics of the SPR? eg if creating and maintaining the SPR roughly doubles the cost of each barrel, and then we can only draw down half of the SPR, then we have already "pre-shocked" ourselves with a 4x price increase. How much does this make sense, vs (eg) just keeping a pile of money around to stabilize the price by guaranteeing a buyer of domestic supply? I don't love price supports but I also don't love big holes in the ground that eat half the oil we put in them
Some quick napkin math gets to roughly 45 cents/bbl/month for ops + amortised cost (in 2025 dollars.)
Given that the oil is mostly bought at low prices and sold at high prices that seems fine?
Good point that we are generally only withdrawing in high price regimes. Still, if we can only pull out half of the oil, the we are closing in on $1/month/barrel, so … that starts to add up. Thanks for inking up a napkin for me :)
Tbh, you have no idea what China's underground storage reserves are, as only 1.4 Mbbl of storage is visible above ground.
Interesting...two questions:
1. What grade of oil is kept in the SPR? Brent, WTI, a mix of both, or something special?
2. What caused that precipitous drop in the SPR shown on the chart after 2020?