That resource pyramid is just insane: undiscovered and unassessed? Without exploration drilling what is their model for the resource? All rocks of similar type have gas? This is most likely in the same vein as the USGS report from 2000 that was using a trivial statistical model to predict vast oil reserves off Greenland. Basically nothing more than wishful thinking.

I remember reading that in Canada there are vast tight gas reserves. The gas in Canada is much harder to extract than the tight gas in the US, but it totals several times the Worlds conventional natural gas reserves. The next great challenge?

There are tight gas resources around the world. Other than the US, I don't believe that much has been done with them, because other natural gas that was easier to extract has been available.

This is an exhibit from the National Petroleum Council's Facing Hard Truths report showing an estimate of the amount in place (not economically recoverable).

I don't know anything about how difficult Canadian tight gas reserves are to extract. US tight gas reserves were considered impossible, before a lot of research was done.

Regarding Canadian resources, AJM Petroleum Consultants from Calgary has good online pdf's (www.ajma.net then click on "about ajm", then "news and events", then "past presentations"). Also try www.rdfuture.com and click on "natural gas" and scroll down to the 5th graph. In general, AJM is predicting very expensive times ahead for Canadians and their cold climate. And the Canadian Society for Unconventional Gas indicates (if memory serves) that coal bed methane could only make up about 10% of the shortfall by 2020. I got to thinking a lot about this in February when windchill temperatures got down to minus 50 celsius just north of Calgary.

If nothing else, unconventional takes a long time to ramp up. Besides the infrastructure for the production, one needs the pipelines to take it to the users.

Some new US pipelines have become available in the last year. I think that is part of the reason for the increase in production.

Gail

I can't find the link to the site I found this info on. As I remember the problem had something to do with the holes bieng smaller than in US basins, and more difficult to fracture. They expected progress to be slow but the potential resource is immense.

The Wiki NG article on Canadian 4.1 Unconventional gas mentions the "Deep Basin" in Alberta as a massive source of tight gas; maybe that's the one? Low permeability. Dave Cohen's piece Canadian Gas - Decline Sets in likely has more info on tight NG.

Hi diss,

Those YTF ("Yet To Find") numbers are based on things like proven reserves per square kilometre in basins of known age and geological character, like the ones shown on the maps in Gail's posting, adjusted for track record in areas already explored. It's true that they are not precise, but that doesn't mean they are just pulled out of the air. A good explorer should be able to quote central estimates and error bars (traditionally P10-P50-P90).

It's not the case that all rocks of similar type have gas, but a given association of mature source rock, reservoir rock and seal ("hydrocarbon system") has a nonzero probability of containing producible hydrocarbon of a given type. That probability can be estimated from drilling history, in the same system or analogous systems if necessary, and the areal extent of the system can be mapped in a variety of ways. Subtract proven reserves and you've got YTF. You can do this with well penetrations or without - in the latter case, the error bars are larger, but not infinite.

If that YTF estimate was done bottom-up, i.e. by aggregating the estimates for a large number of basins and plays, then it could be quite accurate, even if the individual estimates are way off. Something to do with this http://en.wikipedia.org/wiki/Central_limit_theorem

How do you think oil companies decide where to explore, where not to explore, and when to stop exploring in a given basin?

Any chance they'll find some oil by the by?

Find some oil? Sure, all the time. Enough oil? That's a different question, as I'm sure you know.

If they are using actual drilling data then they do have an empirical model to work from. It is not clear from the various articles whether the yet to find reserves are associated with existing fields or are, like for Greenland, pulled out of the air.

Hi Diss,

Undrilled or sparsely-drilled basins are obviously more difficult, and this should be reflected in the P10-P90 range - which may include zero as a lower limit, meaning non-zero probability that the basin is barren for any of a dozen reasons. In the case of Greenland you would be able to work from analogy with the Jurassic and Triassic of the North Sea. It was the same basin before the Atlantic opened up, though the subsequent geological history would be very different - thermal history affecting hydrocarbon generation, and tectonic history affecting trap creation and destruction. You might have some approximate mapping from coastal outcrops and regional seismic.

It isn't a cookie cutter - every basin is unique, and the available data vary widely in type, quantity and quality. So does the amount of time and detail that you can put into the exercise. As you say, the popular press rarely bothers to mention this important fact.

Unconventional is fairly different from conventional with respect to estimation.

With conventional, one is trying to figure out whether there are fields available which have natural gas (or oil) that can be extracted in the conventional manner. These are discrete fields, and the world has been pretty well explored. It is difficult to find them. This is the reason why we are told that we are discovering less oil each year than we are extracting.

With unconventional, we know fairly well the extent of rocks in place which have natural gas trapped inside of it. I showed a map in Figure 7 for tight gas. Some places will have more sweet spots than others, but for the most part the resource is there. It has already in some sense been discovered, but it doesn't yet pass the proven reserve test.

As I understand it, the amount that is included in proven reserves is the amount that is located in areas which can be drained with existing wells, or has been specifically surveyed and tested. The area nearby in the same formation, which we have every reason to believe is similar to what has been fully surveyed, is "undiscovered".

Advance Resources International is a consulting firm that specializes in unconventional natural gas. They have an amazing amount of material up on their website. This is a link to a document that describes how they derive their estimates.

Thanks for the links and the explanation. Reading the second link the undiscovered category is really the uncertainty in existing plays. The unassessed resources, which have a volume comparable to the undiscovered in the pyramid figure, look like a stab in the dark.

Find some oil? Sure, all the time. Enough oil? That's a different question, as I'm sure you know.

Oh yeah, just speculating on finding the odd Black Swan while they turn the countryside into a pincushion.