All this peak oil talk is great, but how do you model and calculate this stuff?

Is there any way I could do it myself?


I'm sorry if I'm posting in the wrong forum, I haven't used a forum much before, I'm kind of new.


Thanks all!

You can go to the modeling forum on this site.
You will see people doing their own models and posting there.

What kind of data are you going to use?

Well, honestly, I'm new the whole thing. Where would you get reliable information?



Well, I'm not completely new to oil in general, just how to calculate peak oil. Its not something I've encountered enough to understand, which would be why I'm here. I'm planning on giving a report to fellow people to raise awareness if I can. I'm convinced its going to happen soon, the problem with this is I need to make models and gather data in order to effectively present it...I think you get the point.


Thanks all!

Well, now THAT is a good question. The countries with the biggest reserves tend to hide how they calculate their numbers, and nearly never revise them, and when they do, it appears semi-random...at times.

Information alone isn't the problem of course.....modeling? Modeling what? I'm certain you can find lots of people around here who think they can model the political decisions of the Saudi royal family...myself, I'm a more than a little suspicious.

I'm looking for best estimates and how they calculated it, and a geological model..


I'm going to be presenting this to a fairly smart group of people, so just simply saying this and that without any back up won't do. I'm not sure where or how I could go about this so thats why I came here..


is there a formula I could use mathematically?

Well, about the only people who have much of a geologically based answer for anything is the USGS, and around here their answers have more detractors than proponents, which is unfortunate because it appears they've got the most amount of time invested, are neither tree huggers nor drillers, and have access to tons of information most others probably don't.

As far as modelling goes, I'm surprised alot more people haven't chimed in by now, there appear to be quite a few methods and ideas for how they model stuff around here.

I might throw in that the very idea of modeling Peak isn't new, and do you really need to go do it all over again yourself? Why not just grab what appears to be the most reasonable method from the archives around here and throw it out there as your best guess from available information?

Make sure you understand the strengths and weaknesses of whatever model you choose so you can defend it of course, as best I can tell anyone with experience in the field is going to take apart most anything you'll find around here anyway.

Fortunately for someone trying to do what you are doing, if you stick to an audience of "believers" you've got a better chance of WHATEVER you present going over well...this end of civilization thing really seems to resonate with alot of people for some reason.

What kind of presential model would be best to use?


I don't care if its pre made per say, but I need a collection of data to do this.

I'm going to be presenting this in a few weeks so...

I'm the wrong guy to ask which way of modeling is best....most of the stuff around here is just trying to fit production data driven by economics and the political will of the Saudi royal family ...while trying hard to ignore economics and the political side effects of Saudi royal family.

Hi 128shot, welcome.

My advice is that modeling is great, but if you can't formulate a good one in time for your presentation, don't worry. It's not the only way to express to intelligent people the importance of Peak Oil. (Some might laugh at me saying this, considering some of my recent posts concerning the difficulty of communicating PO -- but I've learned a lot from my failures.) For instance, you could mention the U.S. peak in 1970 (and Hubbert's correct prediction of it), or that worldwide discovery peaked in the 1960s, or that U.S. discovery peaked all the way back in the 1930s. All three of those tend to be met with surprise and thoughtfulness. Or you could talk about the relative net energy yields of renewables vs. hydrocarbons, that can also raise a few eyebrows. You could even start out with a point about national security, most people aren't bothered by that.

What not to start with: all the ways in which our society and civilization are dependent on oil. Do not mention any hard-lander views about possible global depression or recession, or hydrocarbon-based agriculture or pharmaceuticals, until your audience is already engaged and listening. And for goodness sake, don't mention population or closely related concerns, or you'll see everyone's eyes glaze over right before they start making fun of you and then change the subject to something safer. In my experience this reaction doesn't have much to do with intelligence level, but rather predisposition to what I'll call 'ostrichism.' Just get them interested in the concept of the Peak, and then if they are intelligent, they can connect the dots.

Most everything you will find is actually not modeling. It is curve fitting to empirical formulas using heuristics of various kinds.

If you don't mind getting attacked by professional geologists and petroleum engineers who seem to feel threatened by the approach, I would suggest you look at a real stochastic model for how oil flows through the system starting with discovery data. This basically models all the decisions as 1st-order rate constants.

I have a fresh post up where I use the old standby of the Laplace transform to derive the results in an alternative way.
http://mobjectivist.blogspot.com/2006/0 ... place.html

Am I a Doomer or Math-fanatic? You decide.

For a tenderfoot modeler, I recommend taking a look at the BP Statistical Review of World Energy (on BP.com) for the reserves and annual production data. The data may or may not be really good, but at least it comes from a reputable source, and is consistently calculated every year, so if lucky, it will be off by the same amount consistently.

WHT is right. Most of what these guys are doing is trying to arrive at a mathematical expression that fits the historical data, and then using the same expression to try to predict future data. The main "debate", as it were, is how to select the best mathematical expression. Hubbert used the "logistics curve" for this, but there are a variety of other equations that should be a little better.

reservegrowthrulz is also correct in that the models do not necessarily take into account the actions of humans and/or events that may occur in the future. One need only look at the production between 1975 and 1990 for an example of this.

As a start, using your favorite spreadsheet program, graph out the historical production. For the right hand side of the graph, you can test what would happen if production remains constant for the next few years. What does the production curve look like? How long can this go on before we run out of reserves?

Next, try one where instead of flat, try one where production decreases by 2% per year. What does the curve look like? How long will the reserves last?

It's all about trying to predict the future. It's also kind of addictive, until you reach a certain point of head-scratching, where you try to figure out what it all means. Several people on the board are at that point already.

Anyway it's good practice to learn how to run a spreadsheet and graphing program, so have at it! Report back when you come up with a good one.

The addiction comes about when you basically slide the discovery curves by 20 to 40 years over the production curves and see that they kind of match up. You then get to wondering how that comes about, and, voila, you have a theory. But the hard part is that the theory has to fit together under all circumstances or it will start to collapse under its own weight.

I REALLY wish people would stop this beginner mistake kind of misdirection already. Reserves are not the RESOURCE base, if they were, we would have run out of EVERYTHING about 7 years after ANYTHING was discovered, because if you just track RESERVES through time, it turns out, for as long as there has been reserves, there has nearly always been the same AMOUNT of reserves ( expressed as reserves/rate = some amount of time), even after producing a few billion barrels of oil over the last X amount of time, X amount of production being an order of magnitude higher than what USED to be called reserves...30 years ago.

Reserves are simply a forecast ruled by engineers and accountants who figured out a long time ago that storing future production under ground, behind pipe or undeveloped, was better than investing a few gazillion dollars of capital in a week and then storing everything on the surface waiting 50 years for demand to catch up with their supplies.

Come ON people, learn some of the terminology already before presenting ANY of this stuff to someone who knows the difference between reserves ( SEC, international or make believe ), resources ( defined by whoever you want but at least mention them), EURs or Ultimates, PDP's, PDNP's, PUD's and if you know at least a FEW of the SPE definitions, that would be a good start.

No wonder you are so confused. I only use estimates of discovery volume in the ground. You can only pump out what you can discover. Thinking about reserves as some big tank you keep in the backyard is obviously misguided. I don't think anybody believes that.

If people in the oil industry can't get their act together, that is their problem. As engineers we have one way to measure capacitance, one way to measure mass, one way to count time. If the oilers make up their own measures for obfuscation purposes, then you basically need us unwashed masses keeping an eye out for the deceptions being promulgated.

Like I said before, this is basically a bean-counting exercise and I know that they know how much is under the ground. If they were to spill the beans, we would all be better off.

The analogy would be the weather forecaster who can estimate an average amount of snow that will fall tomorrow given some past historical conditions. However, for whatever agenda he has up his sleeve, he either wants to low-ball the number or high-ball it, thinking that no one really call him on it, because, after all, it is only a prediction. But people do really care, and they will fire the weatherman pronto and instead hire somebody that knows something about Bayes theorem. But the weird thing is that the weatherman clearly wouldn't have an agenda for purposefully getting it wrong, so he always tries to do his best.

So if an oil company was a farm, they would understimate the amount of irrigation needed (and the crops would dry out), they would wait to long to harvest (and the crops would rot), they would not rotate (and the nutrients would leach out of the soil), and when the local bank loan manager would question what happened, they would make up a sob story. And they wouldn't listen to the weatherman, because as they would tell you, "you don't need a weatherman to tell you which way the wind blows".

The whole attitude is about a gnat's eyelash away from believing in The Rapture. You know, God's will and all that jazz.