Shale Reservoirs, do they work, will they spread?
By Andy May
Popular accounts of shale oil and gas reservoirs are often riddled with errors and, even when technically correct, often misleading. As a shale petrophysicist, retired from Devon Energy, I thought I would try and explain, in a non-technical way, how these reservoirs work and why they have been so successful.
We often hear the assertion that shale reservoirs deplete so fast that they cannot be economically viable over the long term and that shale development is a losing “house-of-cards” proposition that will ultimately fail. This slide presentation by Jeremy Leggett is just one example of many. Proponents of solar, wind and other alternative energy sources love to spread such stories, see here for an example. Be careful, not everything in these links is true and they are quite selective in their “facts.” Generally, these nay-sayers rely heavily on industry averages of cash flow or profit and unsubstantiated, controversial and sweeping environment claims to make their point, quite dangerous in a new technology business.
The environmental claims
Many of the environmental claims about shale gas and oil are about methane leaks, because methane is a strong greenhouse gas. However, methane degrades quickly in the environment to CO2 and water by way of a couple of intermediate reactions. According to the EPA, total industry emissions of methane in the U.S. are only 31% of the total in 2016, and they are declining rapidly as technology improves. The largest source of methane emissions are natural wetlands, bacteria in shallow coal seams and termites according to the same EPA source.
We often hear the claim that ground water is contaminated by hydraulic fracturing, yet no verified claims of this have been found in numerous studies. Ground water can be contaminated if a well’s steel casing and cement fails or fluids are spilled on the surface, but these problems are uncommon, local and can be fixed (EPA). Millions of wells have been drilled in the United States and mechanical failures, affecting the ground water, have occurred in very few wells. The rate of mechanical failures in hydraulically fractured wells in the Denver-Julesburg (DJ) Basin of Colorado, has been estimated at 0.06% to 0.15% and none were due to the fracking process itself (Sherwood, et al. 2016). For a discussion of why this is so rare see this post by Barry Stevens. By far, the most common source of methane in ground water is biogenic gas created by bacteria in shallow coal or the soil and in the ground water itself (EPA). Sherwood, et al. note that water wells drilled in the DJ basin in the 1880s produced flammable gas at 400 meters. At that time there were no oil and gas developments in the area.
The business viability claims
Pardon my bluntness, but claims that shale gas and oil production is not economically viable are obvious nonsense. Forty-four percent of Devon Energy‘s reserves, at the end of 2017, were in the Barnett Shale alone. Devon Energy is my old employer and, full disclosure, I still own stock in the company. Thirty-two percent of Devon’s production is from the Barnett and Eagle Ford shales. They have drilled over 5,000 wells into the Barnett Shale since 2002, yet their operating cash flow is up 61% in 2017 and they have raised their quarterly dividend, even though they have been in a tough price environment.
Shale gas and oil exploration, development and production are new technology. Some companies are better at it than others, so using averages across the industry makes no sense in this immature, new business. In the early days of any high technology industry, many companies are created and only a few of the best thrive. How many people remember Apollo Computers or Digital Equipment Corporation? Judging the microcomputer industry based on these companies, or an average of all microcomputer companies in the 1980s, one could easily conclude that microcomputers have no future, but they would be wrong.
Shale is attractive because it is relatively low-risk from an exploration standpoint, but tough because each reservoir is different and the learning curve to make a development profitable is steep and very expensive. The Barnett Shale (see Figure 1 for location) play took years and millions of dollars to figure out. George Mitchell and Mitchell Energy began to work the Barnett Shale problem in 1981 and it did not bear fruit until the early 2000s. Continental Resources drilled many unsuccessful wells into the Bakken in North Dakota before they figured out how to drill a profitable one, and the Bakken is now one of the largest onshore oil fields in the United States, with more than 40 billion barrels of oil-in-place. The USGS estimates that there are over 7 billion barrels of technically recoverable oil in the field. Continental Resources is a very successful company and 60% of its production is from the Bakken Shale.
Overall the shale oil and gas business, at least in the United States, is successful. The USGS estimates that the production of natural gas from shale increased nineteen times in the ten years between 2001 and 2011, in spite of the recession in 2008. Shale production is the main reason the United States is now exporting both natural gas and oil and currently is the largest oil producing country in the world according to David Middleton and the U.S. EIA. It is also one of the principal reasons why oil and gas prices fell after 2008. But this post is not about the economic viability of shale gas and oil, it is about the rocks and how we get oil and gas out of them.
The basics of a shale gas and oil reservoir
A successful shale reservoir is composed of a rich oil or gas source rock, with an adjacent brittle unit that can be hydraulically fractured (“fracked”). The horizontal completion wells will be placed in or very near the brittle rock which is the conduit through which the oil and gas will be produced.
A shale’s defining characteristic is that it is very thinly bedded and fissile, it can also be defined as a fine-grained detrital sedimentary rock. When we say shale in the context of shale oil and gas reservoirs, a more proper term might be mudstone, which is any rock composed mostly of grains smaller than silt size (smaller than four microns across). Here we will use the term shale, but we are referring to rocks with very small grain sizes. We expect, but do not require, that many of the grains will be clay minerals, like illite, kaolinite or chlorite. Typically, oil or gas source rocks are shales or fine-grained carbonates that contain more than 1% organic matter. Once the organic matter is heated to an appropriate temperature and placed under enough pressure it begins to turn into oil or natural gas, depending upon the type of organic matter and the temperature and pressure (see Figure 2).
The generated oil and gas both have a lower density than the organic matter, often called kerogen, that produced it, so once formed it cracks the rock holding it and leaves void space in the kerogen. Shales and fine-grained carbonate source rocks are almost completely impermeable to fluids until this happens (see Figure 3). The native intergranular permeability, from core analysis (Luffel and Guidry 1992) (Luffel, Hopkins and Schettler, SPE 26633 1993), in these reservoirs is around 300 nanoDarcys (10-9), conventional reservoirs have permeabilities of over 10-4 Darcys. Once the escaping hydrocarbons and other geological forces crack (or naturally fracture) the shale the overall permeability goes up one to two orders of magnitude, from around 0.0003 milliDarcys to 0.009 to 0.02 milliDarcys according to perforation inflow test analyses(Rahman, Pooladi-Darvish and Mattar 2005). The level of natural fracturing is an important controlling factor in hydrocarbon producibility (Curtis 2002).
The whole rock unit will not contain organic matter, so one of two things happens once the oil and gas are generated.
If the source rock is adjacent to a permeable water-filled rock layer, such as a sandstone, the oil and gas will migrate into the permeable bed and, since it is less dense than the water in the permeable bed it will migrate up, until it is either trapped in a conventional reservoir or escapes to the surface where it is eaten by bacteria that have evolved to consume oil and gas. The second process, if no adjacent permeable bed is reachable, is to continue cracking the source rock. This causes a buildup of pressure in the rock and it becomes “over-pressured.” Strictly speaking, an over-pressured rock has a pore pressure that exceeds the hydrostatic pressure at the same depth, that is the pressure exerted by a column of water as high as the depth of the rock from the surface. A column of fresh water produces 0.433 psi/foot, so at 8000 feet the “normal” pressure would be 3,464 psi. Any pressure above that is considered over-pressure. Most good shale reservoirs are over-pressured.
Thus, some oil and gas is stored in the source rock as “free” oil and gas in small natural cracks and in larger pores. A second way that shale reservoirs store hydrocarbons, is as adsorbed gas and oil attached to the oil- and gas-wet particles, that is, the organic matter in the rocks (Spears and Jackson 2009). Some reservoirs, such as the Barnett, contain a lot of adsorbed gas and oil (Ambrose, et al. 2010). Adsorbed gas and oil is proportional to the amount of organic matter in the rock (Lewis, et al. 2004). This is because the other particles in the rock are mostly water-wet. Wettability simply means that a rock particle, in the presence of hydrocarbons and water will attract water to its surface if water-wet or hydrocarbons if oil- or gas-wet. When the pore pressure is lowered, say by producing fluid from the rock, some of the adsorbed fluids will be released, this increases well production. Figure 4 illustrates the accessible reservoir volume for a shale gas reservoir.
The structure of a shale reservoir
Shale reservoirs are all different, which means that completion designs are always customized for each reservoir (Rickman, et al. 2008). Most shale reservoirs have very plastic (flexible) rocks with a lot of organic matter (kerogen) and brittle rocks with less kerogen and more carbonate or silica. Generally, the operator will want to drill the horizontal well in the brittle rock, but close to the plastic rock which contains most of the oil and gas. The more brittle rock will drill faster, and it fractures (fracks) well (Jacobi, et al. 2008). Further, the proppant will hold the fractures open better in the harder more brittle rock. In the plastic rock the rock can collapse the fractures by deforming around the proppant.
In Figure 5, the Upper Barnett and the Lower Barnett are more plastic (lower Young’s Modulus) and the central Forestburg and lower Chappel Limestones are brittle (or stiffer, higher Young’s Modulus). These are often good places to put a horizontal well. There are similar “hard” or “brittle” zones in, over or under the Eagle Ford, Haynesville Shale, Bakken Shale and Marcellus Shale.
Figure 6 shows that the Bakken and Three Forks shale reservoirs have a similar structure.
Hydraulic fracturing (“Fracking”)
While micro-fractures exist naturally in shale reservoirs, they do not interconnect enough to form an economic hydrocarbon flow network. For this reason, operators will hydraulically fracture their wells, usually in stages over a three to five-day period. This is illustrated in Figure 7.
Besides fracturing the rock around the wellbore, proppant, which can be sand or ceramic pellets, needs to be placed in each fracture so the fractures don’t close when the well is produced. We also hope the rock is “stiff” enough not to bend and close off the fracture around the proppant. This is illustrated in Figure 8.
The design of an economical well completion procedure for a shale play is the most important component in a successful shale discovery. This is where successful companies stand out and it is where all the scientific and engineering research has been focused over the last 25 years.
Today most operators in the successful shale fields use “slick water” fracks in their wells. Slick water fracks use 98%+ water, 1-1.9% proppant, with a small amount of a friction reducing chemical (usually polyacrylamide), a chemical to lower the water viscosity, minor surfactant (detergent), a small amount of biocide, and additives to reduce corrosion. The advantages of a slick water frack included:
- Lower cost
- Can be pumped at higher pressures, creating more complex hydraulic fractures.
- The water in the frack enters the rock more easily, attaching itself to the water-wet particles and forces the hydrocarbons out.
The details of the slick-water frack are varied for each well, depending upon the characteristics of the rocks. Not only is every shale reservoir different, but each well is also different. This is a learn-as-you-go process and expensive.
As shown in Figure 8, the idea is that the fracture near the wellbore perforation be wide and propped open with abundant proppant. This rock should be reasonably stiff (relatively high Young’s Modulus) and brittle (Jacobi, et al. 2008). Then the fracture will hit the high organic-content and less stiff (low Young’s Modulus) source rock that contains the oil. This will cause the fracturing fluid to slow down, pressure to build, and the fracture will naturally branch out. This puts the water into contact with more rock volume.
Nearly all shale reservoirs contain relatively salty water, the more saline the water the more compact the clay minerals in the shale. The slick water frack is composed of low salinity water, normally lower than the formation salinity. This has two effects on the rock, first it causes the clays to swell slightly, this has the effect of pushing the hydrocarbons into the fractures. Second, the water-wet grains (mostly clays and silica particles) will adsorb the fracturing water, pushing more hydrocarbons into the fractures. These processes, two parts of the same process, account for much of the initial production surge in shale wells.
Every shale is different, some shales, like the Haynesville Shale, have a lot of free gas and are very over-pressured. In these wells, only a small amount of the fracturing water is recovered, and the initial production surge is huge. The water replaces the void space of the produced hydrocarbons quickly, the production is quick, and the wells deplete rapidly. This is one extreme.
In the Barnett Shale the over-pressure is less, so the initial production surge is less. However, the Barnett is a more competent rock and the fractures prop open better. The amount of adsorbed gas is larger than in the Haynesville shale and the release of the adsorbed gas due to the production-caused pressure drop is longer lived and the wells have a longer life. The initial production surge is composed of the free and dissolved gas and oil. The production in the later portion of a shale well’s life is from the release of adsorbed gas. Coal-bed methane production, as another example, is often nearly all adsorbed gas.
The economics of each well is affected by the time it takes to recover the investment in the well and the minerals lease. For this reason, the volume recovered in the initial production surge makes a huge difference. Good Haynesville Shale wells can pay-out in a few weeks, Barnett Shale wells take much longer. All shale plays have sweet spots that are very profitable and areas where the wells struggle to pay for themselves, so mapping the critical petrophysical properties for a good producer is very important. Good operators will use these maps to avoid the sub-economic and marginal parts of shale plays.
Why hasn’t the shale technology spread more quickly?
The shale revolution has taken off in the United States, Argentina and Western Canada. Here we discuss shale oil and gas and exclude oil shales, which have been mined for over one-hundred years in China, Spain, Estonia and some other places for kerogen. Kerogen is not oil or gas but can be converted into them. We are only talking about true oil and gas already formed but trapped in shale.
In the United States and Canada, the plays are being developed on private land where the land owners own the minerals and receive royalties on the production. In most countries, a land owner only owns the upper six feet or so of his land, the minerals and water below that depth are owned by the government. In these countries, the land owner has no incentive to develop the minerals on his land, and every incentive to try and block the development. Leases on government land are rare due to landowner opposition, the strong environmental lobby and for other political reasons.
Shale developments require more wells, roads and construction than conventional oil and gas developments making permitting and regulatory compliance difficult. So, without help from local land owners, the developments often do not happen.
The exception is Argentina in the Vaca Muerta formation in the Neuquén Basin. In Argentina, the mineral rights are owned by the government, but they are pushing the development of their rich Vaca Muerta shale reservoir, which may contain 27 billion barrels of recoverable oil. The Neuquén Basin has very poor infrastructure which makes development expensive and difficult. Further, the government oil company (YPF) lacks the required expertise. So they have invited foreign companies to develop the field and have subsidized unconventional oil and gas prices (source oilprice.com) to entice them to come. The government wants them to build an experienced labor pool and the necessary roads and infrastructure.
Thus, we see the problem. The U.S. and Canada have the technical expertise and an experienced labor pool. They also avoid the political/environmental movement nonsense since landowners own the mineral rights and profit from the shale development. Finally, the U.S. and Canada have the high-quality oil and gas infrastructure and service companies required to do the shale evaluations, engineering, drilling, completions and transportation required. Most other countries have none of this.
Conventional oil and gas expertise is cheap and readily available. Production costs for most conventional onshore oil and gas fields are also cheap, so oil and gas companies can afford to build the infrastructure and develop the fields gratis, just to get the leases. Shale exploration is cheaper than conventional exploration, but the development and production of shale gas and oil is much more expensive. So, for a country to get their shale assets developed they are probably going to have to what Argentina has done and pay some sort of subsidy or guarantee to the operator. This is not going to be politically popular and may be politically impossible for the time being for most governments. The good news? Like all new technologies, shale will get cheaper in the future and the expertise will spread with time. At some point shale oil and gas development will mature and spread around the world. When it does, it will change the world, just like it has changed the United States.
The post is based mostly on my years of experience as a shale petrophysicist and it is intended to be a non-technical layman’s overview of the science. The following recommended articles are for those that want to see the technical details. Collectively, in combination with the links, they cover the same issues as in the post, but in far more detail.
Ambrose, R., R. Hartman, M. Campos, I. Akkutlu, and C. Sondergeld. 2010. “New Pore-scale Considerations for Shale Gas-in-Place Calculations.” SPE Unconventional Gas Conference. Pittsburgh. https://www.onepetro.org/conference-paper/SPE-131772-MS.
Curtis, John. 2002. “Fractured Shale-Gas Systems.” AAPG 86 (11). https://pubs.geoscienceworld.org/aapgbull/article-abstract/86/11/1921/39953/fractured-shale-gas-systems?redirectedFrom=fulltext.
Jacobi, D., M. Gladkikh, B. LeCompte, G. Hursan, F. Mendez, J. Longo, S. Ong, M. Bratovish, G. Patton, and P. Shoemaker. 2008. “Integrated Petrophysical Evaluation of Shale Gas Reservoirs.” CIPC/SPE Gas Technology Symposium. Calgary. https://www.onepetro.org/conference-paper/SPE-114925-MS.
Lewis, Rick, David Ingraham, Marc Pearcy, Jeron Williamson, Walt Sawyer, and Joe Frantz. 2004. “New Evaluation Techniques for Gas Shale Reservoirs.” Schlumberger Reservoir Symposium 2004.http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.455.2453&rep=rep1&type=pdf.
Luffel, D., and F. K. Guidry. 1992. “New Core Analysis Methods for Measuring Reservoir Rock Properties of Devonian Shale.” JPT (SPE) 1184-1191. https://www.onepetro.org/journal-paper/SPE-20571-PA.
Luffel, D., C. Hopkins, and P. Schettler. 1993. “Matrix Permeability Measurement of Gas Productive Shales.” SPE Conference. Houston. https://www.onepetro.org/conference-paper/SPE-26633-MS.
Rahman, N., M. Pooladi-Darvish, and L. Mattar. 2005. “Perforation Inflow Test Analysis.” Canadian Internation Petroleum Conference. Calgary: Canadian Institute of Mining, Metallurgy and Petroleum. https://ihsmarkit.com/pdf/pita-paper_228014110913049832.pdf.
Rickman, Rick, Michael Mullen, James Petre, William Grieser, and Donald Kundert. 2008. “A Practical Use of Shale Petrophysics for Stimulation Design Optimization: All Shale Plays Are Not Clones of the Barnett Shale.” SPE Technical Conference. Denver: SPE. https://www.onepetro.org/conference-paper/SPE-115258-MS.
Sherwood, Owen, Jessica Rogers, Greg Lackey, Troy Burke, Stephen Osborn, and Joseph Ryan. 2016. “Groundwater methane in relation to oil and gas development and shallow coal seams in the Denver-Julesburg Basin of Colorado.” PNAS 113 (30): 8391-8396. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4968736/.
Spears, Russel, and Lance Jackson. 2009. “Development of a Predictive Tool for Estimating Well Performance in Horizontal Shale Gas Wells in the Barnett Shale.” Petrophysics. https://www.onepetro.org/journal-paper/SPWLA-2009-v50n1a1.
via Watts Up With That? https://ift.tt/1Viafi3