Oil and Gas Evaluation Report : Oil & Gas Evaluation Professionals : Obsidian Energy Company : · Seismic Interpretation, Petroleum Engineering

I recently had the opportunity to attend the SPE Annual Technical Conference and Exhibition (ATCE) in Florence, Italy (yes it was very nice). One of the highlights for me was a paper presented by Dr. Lee of Texas A&M University titled A Better Way To Forecast Production From Unconventional Gas Wells. The paper is SPE 134231 authored by Peter P. Valko and W. John Lee, both from Texas A&M University.

In this paper, and another paper authored by John Lee and Rod Sidle, Texas A&M (SPE 130102) a method for forecasting reserves using decline curves is presented. This method, termed “Stretched Exponential Decline” uses a different set of equations than most of us are used to using for forecasting these types of reserves. 

 Historically we have used equations developed by Arps in 1945 which describe three variations of the decline equations for exponential, hyperbolic and harmonic declines; the difference being the value of “b” the “decline exponent.” The general form of the equation is

 q= q_i/(1+bD_it) ^(1/b)

 Exponential decline (straight line on a semi-log graph) occurs when b=0. Hyperbolic decline occurs when b is between 0 and 1 and demonstrates a curved plot on a semi-log graph, like we have seen in shale gas wells during early time production; and harmonic decline occurs in the unique case when b=1.

 In practical experience, hyperbolic declines are often used to forecast tight reservoirs, such as shales, since the shape of the early-time production data can be matched using these equations; however, it is not uncommon for “matched” b factors to be well in excess of 1, which is outside of the parameters described by Arps.  The problem is when the b factor is 1 or greater the Arps equation will approach infinity; which is obviously not possible. In real life most evaluators deal with this problem by placing an arbitrary minimum limit on the production decline and forcing the forecast to an exponential decline late in the well’s life. This arbitrary minimum limit solves the problem of infinite reserves, but it is arbitrary and different evaluators may use different limits.

 Lee et al’s Stretched Exponential Decline takes a different approach. This approach is totally empirical and can be thought of as a sum of a series of individual exponential declines with differing decline rates. In other words, it’s like a given shale well is producing from multiple smaller volumes with each behaving exponentially (heterogeneity). The mathematics are a little more complicated, and definitely outside the scope of this blog, but supposedly Valko has developed software to handle the difficult parts (I have yet to see or use the software).

 The advantages of the Stretched Exponential Decline approach are many; including a bounded EUR (Estimated Ultimate Recovery) and a graphical straight line of recovery potential (rp) versus cumulative production. Experience with the Arps hyperbolic equation is that as more data become available over time for a given well the “matched” b factor is usually reduced from earlier matches suggesting early-time estimates of recoveries may be reduced over time (depending on how the “tail” was handled by the evaluator, as discussed earlier).

 Using decline curves to determine reserves is a very common and important methodology available to the evaluator. This is even more important in the shale-type resource plays where other traditional methods of determining reserves are limited by data availability and our understanding of the production mechanism, and is compounded by the need to determine expected reserves early in the life of a play for business decisions such as leasing and drilling. I don’t know if the Stretched Exponential Decline method will catch on and become the norm, I guess it will depend on the ease of use and whether the economic software providers support it, but I applaud the effort to understand the production mechanism and attempt to create a usable model for the evaluator.

• Email This
• Print
• Comments
• Trackbacks
• Share Link

This article was contributed by Staffan Van Dyke

The objective of this article is to evaluate tight gas sandstones in relation to conventional reservoirs (sandstones/carbonates) as well as unconventional reservoirs (coalbed methane/shale gas), with reference to its constituent petroleum system parameters: source, trap, seal, reservoir properties (porosity and permeability), and time factors (timing of charge and migration). The article indicates significant differences between tight gas sandstones as compared to coalbed methane and/or gas shales. A thorough evaluation of the geological evidence studied for this article indicates that tight gas sandstones, as a reservoir, are closer to conventional type reservoirs than they are to unconventional type reservoirs, such as coalbed methane and/or gas shales.

Utilizing the framework described in this paper, tight gas sandstone reservoirs should then be considered as a sub-type category within the overall conventional reservoir definition, as the majority of its geological properties fall within this definition, and not that of an unconventional reservoir – note: the suggestions laid out in this article STRICTLY refer to the geological parameters of these reservoirs and NOT their engineering parameters (which are still very clearly considered “unconventional”).

Under this definition just laid out, the characterization of tight gas sandstones as an unconventional reservoir is simply inappropriate, as the geological setting / petroleum system / etc., as compared to coalbed methane and shale gas, are very different in their most basic geological constituents, making the comparison (and, hence, the argument that they are indeed very different from one another when viewed in a geological sense). Tight gas sandstones are simply reservoir rocks, whereas coal and shale are considered to be both the source and the reservoir rock.

Unconventional reservoirs are ones that cannot be produced at economic flow rates or they do not produce at economic volumes without the assistance from massive stimulation treatments, such as hydraulic fracturing (fracking) or other special recovery processes and technologies, such as steam injection (these are known as “secondary” and/or “tertiary” recovery techniques). Typically unconventional reservoirs have been described as: tight gas sands, coalbed methane, and gas shales (Holditch, 2003 and 2006). However it is an economic and reservoir engineering definition and does not take into account the geological processes behind the deposition of said deposits.

It is also important to understand that a conventional (sandstone/carbonate) reservoir with low natural pressure that depletes very quickly (in the order of weeks to months) that requires artificial hydrocarbon recovery techniques to maintain or increase its economic viability, is very nearly the exact definition of an unconventional reservoir, as the one given above. However, such reservoirs are still categorized as conventional in the geological sense.

On the other hand, since tight gas sandstones must be artificially stimulated (fracked) in order to produce its gas, it would only seem natural to place this reservoir criterion in the “unconventional reservoir” category.

Comparison of conventional and unconventional reservoirs

In the United States, the tight gas sandstone definition is applied to reservoirs with less than 0.1 mD of permeability (Meckel and Thomasson, 2008). Our investigation indicates that tight gas sandstones have significantly different characteristics in comparison to coal bed methane and shale gas. They are:

1. Tight gas sandstones act purely as a reservoir, whereas coalbeds and shales act not only as their own source rocks, but as well as their own reservoirs;

2. Shanley et al (2004) found that the low permeability reservoirs in the Greater Green River Basin of southwest Wyoming were not part of a continuous type gas accumulation but were low permeability rocks in conventional structural, stratigraphic, and/or combination traps. Earlier, Berry (1959) and Hill et al (1961), proposed that in the San Juan Basin, the gas within the sandstone reservoir was localized in a potentiometric sink associated with down-dip flow of water. In other words, it is a hydrodynamic type trap, thus, much more like the conventional trap settings found in conventional reservoirs;

3. Gas migrates into tight sandstones from the nearby source rock and the charged gas may be housed within the reservoir due to high capillary pressure conditions by virtue of low porosity and permeability, and up-dip presence of water due to regional or local hydrodynamic conditions, whereas in coal and shale gas, it is adsorbed into the matrix of organic matter (Bustin and others, 2009);

4. Many conventional reservoirs are porous and permeable but do not have enough primary energy to support hydrocarbon production unaided at an economic level, but are still categorized as conventional reservoirs. According to the unconventional reservoir definition given above, this quality should then define these reservoirs as unconventional, primarily because enhanced recovery techniques are required for them to be economically producible. Similarly, tight gas sandstone reservoirs need enhanced recovery techniques like fracturing, flooding, and
acidization to make them economically viable. However, instead of categorizing these low primary-energy conventional reservoirs as unconventional, it is the authors’ opinion that they should remain classified as conventional reservoirs, and that tight gas sandstones should be classified as a sub-type within the overall conventional reservoir petroleum system;

5. The only correlatable property of tight sandstones to coal and shale is their low porosity and permeability similarity, unlike the higher porosities and permeabilities typically seen in conventional sandstone / carbonate reservoirs. The geological aspects discussed above suggest that tight gas sandstone as a reservoir is closer to conventional reservoirs (sandstone / carbonates) than to coalbed methane and shale gas reservoirs. Table 1 summarizes the petroleum system and other parameters with respect to tight gas sandstones, coalbed methane, shale gas, and conventional reservoirs to elucidate the similarities between these reservoir types.

Conclusion

Evaluation of the above geological aspects suggests that tight gas sandstones, as a reservoir, are closer to conventional type reservoirs than to unconventional type reservoirs, like coalbed methane and shale gas. It is clear that tight gas sandstones act simply as a reservoir, whereas coal and shale act as a source rock as well as a reservoir for the gas. Tight sandstones may become a hydrocarbon reservoir only when a potential source rock is available within the basin, or a nearby region, capable of charging the reservoir. Utilizing the framework described in this paper, tight gas sandstone reservoirs should be considered as a sub-type conventional reservoir, as the majority of its geological and petroleum system parameters fall within this definition, and not that of an unconventional reservoir.
 

• Email This
• Print
• Comments
• Trackbacks
• Share Link