by Vaclav Smil
This technique was patented in 1949 (Howard, 1949). The Halliburton Oil Well Cementing Company (Howco) became the exclusive licensee of the Hydrafrac process, and it used it for the first time on March 17, 1949 (Green, 2014). On that day, Howco performed two commercial fracturing treatments—in Stephens County, Oklahoma (costing $900), and in Archer County, Texas (costing $1000)—and in the first year, it treated 332 wells, boosting their production by an average of 75%. The technique was rapidly adopted by many producers and by the mid-1950s more than 3,000 wells were treated per month, and Montgomery and Smith (2010) estimate that between 1949 and 2010 the industry completed some 2.5 million fracturing treatments used for as many as 60% of all completed wells. The key goal is to create durable hydraulic fractions whose highly conductive flow paths would remain effective over decades of production (Vincent and Besler, 2013).
By 1990, both horizontal drilling and hydraulic fracturing were thus fully commercial procedures used overwhelmingly by oil industry to boost the productivity of crude oil extraction. How these techniques were adopted, adapted, and diffused by the US natural gas producers and deployed to extract natural gas from the country’s extensive shale formations is complex, slowly unfolding story recounted by one of its protagonists (Bowker, 2003; Steward, 2007), described in books looking at the unfolding gas revolution (Levi, 2013; Zuckerman, 2013; Gold, 2014) and analyzed from historical and policy perspective (Wang and Krupnick, 2013).
Its beginnings go to the late 1970s, to government tax incentives to spur the development of nonconventional gas resources; its important components included several federally funded research programs (above all the Eastern Gas Shales Project at Morgantown Energy Research Center that examined the behavior of underground fracturing and brought advances in directional drilling) and continued advances in new mapping techniques and 3D imaging (that made it possible to pinpoint richest shale concentrations). Massive hydraulic fracturing was demonstrated for the first time in 1977 as a part of the Department of Energy project, the first horizontal well in Devonian shale was completed in 1986, but decisive breakthroughs toward commercial shale gas extraction came only during the 1990s.
George P. Mitchell, the founder of Mitchell Energy & Development, assembled a small team of experts and kept on defying persistent skeptics in his pioneering search for rewarding gas extraction from Texas Barnett shale. Mitchell’s team took advantage of accumulating know-how coming from federally supported research, their first horizontal well in 1991 was subsidized by the Gas Research Institute, but it took large financial risks during its long pursuit of commercial gas from shales. In 1997, the company introduced a new, less expensive liquid formula containing less (and cheaper) sand and a gelling agent from guar bean.
6.1.1 American Shale Gas Extraction
Once the potential for economic extraction of the Barnett shale became evident, Devon Energy acquired Mitchell Energy & Development in January 2002 and began to combine its experience in horizontal drilling with water-based hydraulic fracturing: in 2002, there were 2,083 vertical wells in Barnett shale, and 5 years later, they were 8,960 wells, of which 55% (nearly 5,000) were horizontal (Brackett, 2008). By 2007, Barnett shale was still the dominant producer of the new resource, but its success led to the reassessment of other shale formations, and in 2009 came the news about the world’s largest supergiant natural gas field—in Appalachia (Figure 6.2).
Figure 6.2 US shale basins.
Most of this American region contains Devonian shales (they extend from south-central New York to southern Virginia and from eastern Ohio to northeastern Pennsylvania and are also underneath the eastern part of Lake Erie) that have been known for their kerogen content and hydrocarbon potential, but in 2002 the USGS put the technically recoverable volume of Marcellus shale gas in the Appalachian Basin at just 56.6 Gm3 with additional 14 Mt of natural gas liquids (USGS [United States Geological Survey], 2011). Six years later, Engelder and Lash concluded that “the Marcellus Shale weighs in with more than 500 trillion cuft of gas in-place spread over a four state area. Continuous natural gas accumulations such as the Barnett Shale produce more than 10% of the gas in-place, which when applied to the Marcellus Shale, translates to a resource that will return 50 Tcf in time” (Engelder and Lash, 2008).
In standard units, that means almost 15 Tm3 of gas in place and roughly 1.5 Tm3 of ultimately recoverable volume, 25 times the 2002 USGS estimate. Because Engelder did not have enough public data to define a Marcellus decline curve, he used pro forma decline rate published in 2008 Chesapeake report in order to estimate ultimate recovery from Marcellus; he also assumed that 70% of the sections in each county are accessible and that the wells have 80 acre spacing and ended up with 50% probability of 13.8 Tm3 of ultimate yield (Engelder, 2009). That was equal to just over 20 years of the country’s annual natural gas consumption in 2009. Following Engelder and Lash, the USGS had also raised its estimate of recoverable Marcellus gas to an even higher total of 2.3 Tm3 (USGS, 2011), more than 40 times its 2002 total: I will return to these large discrepancies in estimating ultimately recoverable gas volumes in the book’s last chapter when I will try to assess how far can natural gas go.
Following the Barnett breakthrough, other formations began to show rapid extraction increases. Fayetteville shale (part of Arkoma basin in Arkansas) began to take off in 2008, and by 2013, the gas output had roughly sextupled. Haynesville shale (underlying parts of southwestern Arkansas, northwest Louisiana, and East Texas) was the next one to take off, with average monthly extraction rising to more than 60% of Louisiana’s gas output by the spring of 2010 after just 3 years of development (but its output declined by 27% between 2012 and 2013). Starting in 2010, Marcellus shale matched that rapid rise. Eagle Ford shale (extending in an arc from the Mexican border northeastward south of San Antonio), Woodford shale (underlying virtually the entire Oklahoma), and Bakken shale have been the other notable contributors.
Bakken shale (part of the Williston Basin centered on western North Dakota and extending to Manitoba, Saskatchewan, and Montana) has been North America’s latest addition to large formations producing both oil and gas. In North Dakota, oil extraction (now higher than in Alaska and California and second only to Texas) by horizontal drilling and fracturing has been accompanied by so much natural gas that (in the absence of adequate pipeline capacity) large volumes of it had to be flared: by the end of 2011, more than one-third of all gas produced in North Dakota was flared or not marketed (USEIA, 2011).
Data on new-well gas production per rig show substantial differences among major shale basins, both in daily rates and in trends between 2007 and 2014 (USEIA, 2014j). In the Permian Basin, average performance has actually declined by half to about 3,000 cuft/day; in Eagle Ford shale, it has remained steady at around 1,200 cuft/day; in Niobrara, it has been up and down; Bakken has seen steady but slow rise to less than 600 cuft/day; and Haynesville and Marcellus have been the star performers: in the former formation, the mean had quintupled between 2007 and 2014 to more than 5,000 cuft, while Marcellus has seen an order-of-magnitude gain, from less than 600 to more than 6,000 cuft/day. In aggregate, in the year 2000, shale gas supplied just 1.6% of the total US natural gas production; in 2005, it was 4.1%; in 2010, it was 23.1%; and in 2013, it reached 40%. This unprecedented ascendance of a new energy resource has engendered both exaggerated hopes and, this being America of hyperbolic claims that need not be based on any realities, also near apocalyptic warnings.
The development of shale hydrocarbons has resulted in local and regional impacts well known from previous cases of resource booms as well in some consequences specific to the industry. Sudden influx of new workers (dominated by single males) creates shortages of suitable housing, drives increases in real estate prices, and overtaxes planning and zoning capacities of smaller settlements that are overwhelmed by new permitting requests. Relatively high earnings in the industry increase the cost of business for other sectors, create labor shortages in low-paid
service establishments, and prevent economic diversification because intensive resource development lowers the chance of attracting concurrently other investments; the latter reality only worsens longer-term economic prospects once the initial expansion stage is over, and depopulation rate may be higher than in places with no comparable resource development.
Environmental impacts include some well-known phenomena as well as some new annoyances and damages. As with any kind of hydrocarbon exploration, drilling, and production, there will be local surface land disturbances, destruction of plant cover, and habitat fragmentation. Good news is that shale gas drilling is now often done from space-saving multiwell pad sites taking up 1.6–2.0 ha during the fracturing phase but only 0.4–0.6 ha after site restoration (NYSDEC [New York State Department of Environmental Conservation], 2009; Figure 6.3). Certainly, the most obvious change specific to HVHF is an enormous increase in truck traffic required to deliver fracking fluids. Inevitably, this leads to road congestion on two- or single-lane rural roads, road deterioration (to the point of impassability for unpaved roads), rising road maintenance costs, reduced quality of life (as inhabitants of formerly isolated houses near a rural road suddenly see hundreds of large truck going down their formerly quiet lane), and, most unfortunately, more fatal accidents (Begos and Fahey, 2014).
Figure 6.3 Shale gas drilling site in Pennsylvania.
© Corbis.
While air and noise pollution caused by heavy tanker truck traffic and high-pressure fracturing is only temporarily disruptive, concerns about potential degradation of groundwater quality are much longer lasting, and, not surprisingly, they have attracted a great deal of public attention and protests (Kharaka et al., 2013). Antifracking movement had coalesced first around the concerns about fracking liquids contaminating local aquifers and about drinking water laden with pollutants that could be set on fire when it comes out of a faucet (this became an iconic image of fracking gone awry). Only later came concerns about longer-term effects of air pollution from escaping gases and the increased potential for localized earthquakes. As with all similar movements, antifracking activists form a heterogeneous group whose members range from professional environmental protesters opposed to virtually any resource developments to people genuinely concerned about fracking wells few hundred meters upwind from their backyards, schools, or playfields.
As expected, the cause has attracted its share of celebrities, including not only such veterans of assorted causes as Yoko Ono—whose measured assessment is that “Fracking kills. And it doesn’t just kill us, it kills the land, nature and eventually the whole world”—and Robert Redford but also Alec Baldwin and an Iron Chef Mario Batali (Begos and Peltz, 2013). By September 2013, the antifracturing sentiment reached even to Dallas where the city council rescinded leases granted previously to Trinity East Energy and where at the public hearings the opponents outnumbered those who favored the development, citing increased cancer risk (disproved by actual statistics) at a Colorado fracking site (Ginsberg, 2013).
And Bamberger and Oswald (2014) believe that large-scale fracking has many public health implications, particularly for animals, children, and oil and gas workers. They also compared many people affected by the industry to victims of rape because they are powerless and at the complete mercy of forces beyond their control. Abrams (2014) amplified their claims by publishing her interview with the authors under the title “Fracking’s untold health threat: How toxic contamination is destroying lives.” Some of these harmful exposures have led to successful lawsuits. On April 22, 2014, a family in Dallas County won $3 million in the first Texas case verdict that found a fracking operator (Aruba Petroleum) guilty of fouling the plaintiff’s ranch property, their home, and quality of life and sickening them and their pets and livestock (Matthews & Associates, 2014).
Opposition has been also considerable in New York State, where the movement began with seeking a permanent ban on any drilling in the city’s watershed and where New Yorkers Against Fracking seek to make the entire state off-limits because they consider the activity to be “in the same category as smoking… . The only way to make smoking safe is to not smoke” (Navarro, 2013). Statewide ban was announced in 2014. I will review water requirements of HVHF and the complex evidence regarding drinking water contamination and wastewater disposal in the next chapter.
6.1.2 Shales outside the United States
While the disputes about the consequences of fracking continue in the United States, environmental concerns have been a key reason why several European countries—including France, the Netherlands, Czech Republic, and Bulgaria—enacted preemptive bans (or at least temporary moratoria) on shale fracking. To be sure, there are other factors that militate against US-like European embrace of natural gas. Perhaps none as important as the fact that, unlike in the United States where mineral rights come with the land ownership, the continent’s landowners do not own the rights to develop resources under their land, a reality that does not give them any incentive to get involved in potentially very disruptive extractive activities. Other important differences are a much more flexible access to pipeline capacity and, of course, much lower population density in many of America’s major shale regions and hence a reduced probability of NIMBY conflicts.
This leaves plenty of other countries with the opportunity to develop a new energy resource because hydrocarbon-bearing shales (with organic content of 2% and more) are among the world’s most commonly encountered formations. Besides the United States, other large nations with extensive shale formations include Canada, Brazil, Argentina, Russia (in Western and Central Siberia), Algeria, South Africa, Pakistan, China, and Australia. The first worldwide resource assessment concentrated on 32 countries and only on the regions known to contain within their shale basins the higher-quality prospective areas (hence generally biasing the totals in conservative manner); it included two key judgments regarding the likelihood of potential gas flow for commercial development and an expectation of the extent to which such prospective areas within each shale gas basin and formation will be developed in the foreseeable future (Kuuskraa et al., 2011).
This assessment ended with the total of 623 Tm3 of gas in place and with 163 Tm3 of technically recoverable resources. China’s recoverable volume was put slightly ahead of the United States, with Argentina, Mexico, and South Africa constituting the top five nations. Uncertainty of these estimates is best illustrated by a reassessment of recoverable gas issued by the same organization (Advanced Resources International) just 2 years later (Kuuskraa, 2013). Norway’s recoverable resources went from 2.3 Tm3 to zero due to disappointing results from three wells drilled by Royal Dutch Shell in Sweden’s Alum shale, a similar nearby formation. In a similar case of a near-total downgrading in 2014, the USEIA cut its estimate of recoverable oil from California’s Monterey shale by 96% compared to its initial claim made in 2011 (Reuters, 2014): clearly, more large reappraisals, and indeed eliminations of technically recoverable shale hydrocarbons, must be expected as we learn more about specific formations.
In 2013, Kuuskraa cut Libya’s estimated shale gas volume by nearly 60%, total for France was reduced by 25%, and those for South Africa and Mexico went down by 20% (Kuuskraa, 2013). At the same time, newly assessed basins in Ukraine and Algeria tripled their 2011 totals and added nearly 50% to the Canadian total (as already noted mostly in the Montney formation), and these additions resulted in an overall 35% increase of technically recoverable shale gas worldwide to about 220 Tm3 or about 7.7 ZJ. That is nearly 20% more than the BP’s (2014a) latest global estimate of conventional natural gas resource at the end of 2013 (185.7 Tm3) and nearly four times the total energy (1.94 ZJ in 335 billion barrels) in recoverable shale oil. But this, too, is only a temporary value, and we will get a more realistic understanding of what is really technically recoverable only after every major shale basin will be tested by a significant number of exploratory wells.
Reduction of shale gas recovery for China was minor (by
<10%) but large enough to put the United States in the first place—but the Chinese situation is a near-perfect example of complexities and specificities that must be considered when assessing realistic chances of shale gas development rather than producing generic theoretical estimates based on the extent of shale-bearing basins and their organic content (Chang and Strahl, 2012; Tollefson, 2013). Although China had originally plans for extracting at least 60 Gm3 (and up to 80 Gm3) of shale gas by 2020, it has had virtually no experience in applying and adjusting the two key constituent techniques of horizontal drilling and hydraulic fracturing and limited number of people with requisite expertise. No less importantly, Chinese shales lie deeper than in major US formations, are more scattered, have more fractures and faults, and are inadequately mapped, and hence, they will be almost certainly more costly to extract. Not surprisingly, in August 2014, Chinese production goal for shale gas in 2020 was cut by half to 30 Gm3 (Shale Gas International, 2014).
Water availability is another major consideration: Sichuan Basin has plenty of precipitation, but it is also China’s most populous province with huge water demand for irrigation, industries, and megacities, while western shale basins (Tarim, Junggar) are arid and already experience serious shortages of water (Marsters, 2013). And institutional and entrepreneurial barriers (including state dominance of oil and gas industry, absence of experienced, risk-taking small enterprises, willingness to deal with failure, an almost inevitable outcome of pioneering new production methods in difficult conditions) should not be forgotten.
