What are oil and gas reserves?
When you think of oil reserves, and I’m sure it’s often and with great passion, do you imagine this… or this? Well let’s face it, odds are you never put more than a second’s thought into it. Maybe you’re the kind of person who reads about something like that in the news, and it sounds technical, and seems like a big deal but you have no idea what they’re talking about, and you feel like kind of dumb about the whole thing so you file it away in the little dumpster at the back of your head, and resign yourself to never understanding the concept, though you know it might be kind of important, but the world will go on if you don’t bother looking into it, so you kind of say ah screw it…
Oil and gas reserves are a topic that seems to generate that sort of attitude. That’s a big mistake, on a few levels. First of all, you’re not so dumb or disinterested that you can’t understand the concept, or you probably wouldn’t be reading energy blogs. Second, once you see what goes into a reserves calculation you may be wide-eyed and spooked by the deranged processes that lead to a number you’ll hear in the news. Third, it’s actually kind of important if you’re interested in energy topics. There are some pretty wild-ass numbers that get thrown around as energy reserves estimates, and some pretty big decisions are made based on them. Maybe that’s not a good idea. While not outright pot heads like geologists/geophysicists, reserves estimators use an assortment of techniques not inconsistent with substance abuse, where they calculate some variables to several decimal places (because they have data) and take the wildest guesses at others (the variability of which bails them out when the numbers inevitably turn out wrong).
To answer the little quiz above, believe it or not when you read about “Saudi Arabia’s 260 billion barrels” of reserves, or some such, it’s the rock they’re talking about, not the tanks. So when you hear about reserves, it helps to keep in mind that the numbers you hear relate to a pretty abstract concept, such as how much oil is in that chunk of rock and how much of it can be extracted from it, as opposed to something you can sit there and admire like a big oil tank.
I did struggle tirelessly, upwards of 15 minutes, to come up with a useful analogy to illustrate more clearly the basic and almost loveable goofiness of reserves estimation before giving up in despair. There is nothing quite like it. And in the big scheme of things these numbers have become extremely important with respect to various national energy policies and general war aspirations, and there should be a bit more attention paid to just what these estimates are made of. (And for purposes of this discussion, I’m dropping references to “oil and gas”, they’re similar for the purpose of this discussion, so I’ll just talk about oil reserves because oil is so sweet to type compared to gas, which over time laterally strains my left index finger indescribably. And I wonder why I’m no longer welcome on the drilling rig floor…)
In an earlier post I’ve gone over how difficult it is to find oil and gas in general, and even with a lot of wells and clues, new drilling activity can turn up total surprises. A well can be drilled midway between two good ones and turn out pathetic. Or maybe way better than expected. There can be a lot of variability within a small area even within known oil fields. Over time, and with enough wells in a field, you can get an idea of what the average well might produce. Engineers then try to estimate how long these wells will produce for and at what rate. The total production over the life of the well is called the well’s reserves.
Before getting into reserves incantations, it’s best to first talk about original oil in place, or OOIP, which is an estimate of the total amount of oil contained in one of the underground structures that geologists look for. Oil-bearing zones don’t go on forever; there are always boundaries, though some oil-bearing zones can be really large (such as Ghawar in Saudi Arabia, which is something like 10 by 60 miles of more or less continuous reservoir). By and large, an oil-bearing reservoir will be quite spread out but thin, like a pancake or possibly a snake or some weird combination thereof. If you look at the picture of the rock above, it’s like a microcosm of what an oil reservoir might look like underground (for example, the black part might be a typical reservoir).
OOIP is a pretty hare-brained thing to try to calculate with any precision because the amount of data known about the size of the reservoir is so ridiculously small. If an oil reservoir covers 2 square miles, which by some magical interlude let’s assume you could ever know, and there are 10 wells drilled into it, that means there are 10 6-inch holes drilled over two square miles. These are the only reference points, like ten straws stuck into a football field. Can you make a reasonable assumption about the overall state of the turf of a football field from looking at the core inside those 10 straws? Reservoir engineers think so…
When a well is drilled it either pierces the reservoir and provides information, or misses it entirely and at least tells you where it’s not (that’s part of the optimism inherent in the business – you can drill a dry hole and waste a million dollars and the geologist will say, through a cloud of pot smoke, hey good news, we know where the reservoir boundary is). With enough wells, you can get a rough idea of where the edges are. For really good reservoirs, over time many wells will be drilled and a fair idea of the overall area might be determined.
Big deal. To calculate OOIP, you need to know a few other things: how thick the reservoir is, how porous it is, and how uniform is it across the area – just like a beach can be rocky in some places and wonderfully sandy in others, an underground reservoir can vary greatly. And all this info has to be gleaned from the few six-inch diameter holes poked into the landscape. True, the drilled wells will tell you a lot, but still it’s pretty meagre. Engineers then take those data points and estimate the rest, and come with some grand number like 500 million barrels of oil in place. If they were being completely forthright, they would add “plus or minus 75%” after the number, but their general audience (including some conservative folks such as bankers, I shit you not; the same people who require from small business owners the craziest cash flow forecasts down to the last nickel) actually prefer a firm number, even if it is preposterous. Of course, banks use that number (because it determines how much money they will lend to the company) and then slash it in half, and use that as a starting point. So in the end even the target audience doesn’t trust the number at all.
Anyway, the spreadsheet guys come up with a number for how much OOIP there is, and then set about determining how much of it can be successfully removed. This is where it gets really funny. Most oil in an oil reservoir can’t be recovered, a really good recovery rate might be 50% and that can take decades (and of course “50%” is a meaningless number anyway, because it assumes you knew how much oil was there in the first place…but for sake of getting anywhere, let’s accept that convention for now). Most reservoirs have a 5-20% recovery rate (same problem, 5% of what…), without getting into fancy enhanced recovery techniques like waterfloods that force oil out. (That’s a topic called enhanced oil recovery, good for another day…)
So engineers look at wells that are drilled, at how much they are producing, and decide how long the wells will produce and at what rate. This is a reasonable way to estimate a volume if you are, for example, pouring sand out of a bucket. But with the number of variables and the time frame involved, estimating total ultimate production from a well requires one to make some fairly spectacular leaps of assumptions, and the mental strength to deliver these numbers to stone faced investors, bankers, boards of directors, etc. with a perfectly straight face. It has been common over the past 10 years to hear natural gas companies drill horizontal multi-stage frac wells in parts of the US and, after watching production for 2-3 years, declare that the wells to have a 60 year reserve life. And people just lapped that stuff up like beer. (In case you don’t share my incredulity, at a recent conference a rep from one of the world’s largest fracking company admitted that up to 60% of fracks “didn’t actually work.” If you were a car manufacturer and went home happy at the end of the day because you had 40% of the wheels on right, wouldn’t that be of mild concern to anyone?)
And finally, in order to determine how much oil might ultimately be recovered, the protagonist must make a guess as to what commodity prices will be over the life of the well. This is because reserves are actually “economic reserves”, meaning that the price of oil must be high enough to warrant producing the oil. For example, if oil is expected to be worth $10,000 per barrel, as an extreme example, a lot more of the OOIP will be considered recoverable because people will go to far greater lengths to get at it. It’s like the difference between throwing ten dimes and ten $100 dollar bills into a tank of sewage: how many of each are “recoverable” if you have to rent a scuba suit to get them out? (And why are my analogies always so disgusting? Yet another mystery.)
It sort of makes sense that recovery rates will vary depending on commodity prices, then but again the process becomes silly when predicting oil (or natural gas) prices 60 years out. And when you’re talking about the reserves a well can produce over that long of a lifespan, the price makes a huge difference – if prices drop by 50 percent in ten years and the well can’t produce economically, you would “lose” 50 years worth of production/reserves, and suddenly the headline isn’t so rosy anymore. One final note on reserves craziness before getting to the punch line: Back in the 1970s, OPEC members began squabbling amongst themselves about how much oil to produce in total; if they all produced too much, the price of oil would crater and all would lose. So they determined quotas based on each member country’s oil reserves numbers. Over the period that this process was going on, each country realized hey, if my reserves are higher I can produce more, and the inevitable happened. In the mid 1980s, for example, Saudi Arabia increased its reserves by some miracle from 150 billion barrels to 250 billion. Other OPEC countries managed the same astonishing progress, obviously and utterly unhinged from reality. (Saudi Arabia still claims to have 260 billion barrels of reserves 30 years later, a remarkable feat considering they produce billions of barrels every year.) So when you read about the huge oil reserves that OPEC yet has, remember that a lot of them were added in about 15 seconds by changing a number in a spreadsheet. Trust me, solar and wind power can’t get here fast enough.
One more final note about something you might read in the paper: the petroleum industry makes a big deal splitting reserves into proved, probable, possible, legendary, mythical, etc. Don’t pay any attention to these classifications whatsoever. They are simply changes in the degree of confidence that the calculator has in his numbers, which to their credit is very little (Proved reserves=”I’m pretty sure they’re there and recoverable”; Probable reserves=”I think they’re recoverable”; Possible reserves=”Well, there might be oil there and if there is someone is probably going to be smart enough to get it out one day”, etc. )
Sorry it’s a long post, hopefully you’re still awake. I’m getting to the home stretch. The scary part is that these reserve numbers are taken seriously enough that they become fact in the public debates and lead to policy decisions at the national level. Anyway, a reserves calculation includes a bunch of near-total guesses, all multiplied together, and presented as some sort of concrete basis to make decisions on. No wonder banks take the best estimate and automatically cut it in half. It would be funny if it wasn’t scary. We listen to OPEC and give the credibility because they tell us these reserve numbers, and we believe them. We modify national industrial policy (in the US especially) because some zealous companies tell us their new high-tech wells will produce for more than half a century, which is like looking outside during a rainstorm and predicting apocalyptic floods over the next 10 years because, well, just go look for yourself.
Soooo….just whining about it doesn’t do anyone any good…this website is supposed to be about positive and non-biased energy. So what’s the solution? Well, I don’t have one. Estimating reserves does have value for a number of purposes, for banks to lend money to producers, for governments to formulate policy, but care must be taken not to take these numbers seriously. It is not wise at all to take reserves numbers at face value in the media. Maybe that’s what governments do, I have no idea…but when you read in the news that we have 100 years of reserves, or some outlandish numbers for a relatively new producing area, don’t believe all you hear.
What a mess. How you will sleep tonight I’ll never know.