The majority of FracTracker’s posts are generally considered articles. These may include analysis around data, embedded maps, summaries of partner collaborations, highlights of a publication or project, guest posts, etc.

DOGGR
Kyle Ferrar, MPH

Literally Millions of Failing, Abandoned Wells

/by

By Kyle Ferrar, Western Program Coordinator, FracTracker Alliance

In California’s Central Valley and along the South Coast, there are many communities littered with abandoned oil and gas wells, buried underground.

Many have had homes, buildings, or public parks built over top of them. Some of them were never plugged, and many of those that were plugged have since failed and are leaking oil, natural gas, and toxic formation waters (water from the geologic layer being tapped for oil and gas). Yet this issue has been largely ignored. Oil and gas wells continue to be permitted without consideration for failing and failed plugged wells. When leaking wells are found, often nothing is done to fix the issue.

As a result, greenhouse gases escape into the atmosphere and present an explosion risk for homes built over top of them. Groundwater, including sources of drinking water, is known to be impacted by abandoned wells in California, yet resources are not being used to track groundwater contamination.

Abandoned wells: plugged and orphaned

The term “abandoned” typically refers to wells that have been taken out of production. At the end of their lifetime, wells may be properly abandoned by operators such as Chevron and Shell or they may be orphaned.

When operators properly abandon wells, they plug them with cement to prevent oil, natural gas, and salty, toxic formation brine from escaping the geological formation that was tapped for production. Properly plugging a well helps prevent groundwater contamination and further air quality degradation from the well. The well-site at the surface may also be regraded to an ecological environment similar to its original state.

Wells that are improperly abandoned are either plugged incorrectly or are “orphaned” by their operators. When wells are orphaned, the financial liability for plugging the well and the environmental cleanup falls on the state, and therefore, the taxpayers.

You don’t see them?

In California’s Central Valley and South Coast abandoned wells are everywhere. Below churches, schools, homes, they even under the sidewalks in downtown Los Angeles!

FracTracker Alliance and Earthworks recently spent time in Los Angeles with an infrared camera that shows methane and volatile organic compound (VOC) emissions. We visited several active neighborhood drilling sites and filmed plumes of toxic and carcinogenic VOCs floating over the walls of well-pads and into the surrounding neighborhoods. We also visited sites where abandoned, plugged wells had failed.

In the video below, we are standing on Wilshire Blvd in LA’s Miracle Mile District. An undocumented abandoned well under the sidewalk leaks toxic and carcinogenic VOCs through the cracks in the pavement as mothers push their children in walkers through the plume. This is just one case of many that the state is not able to address.

California regulatory data shows that there are 122,466 plugged wells in the state, as shown below in the map below. Determining how many of them are orphaned or improperly plugged is difficult, but we can come up with an estimate based on the wells’ ages.

While there are no available data on the dates that wells were plugged, there are data on “spud dates,” the date when operators begin drilling into the ground. Of the 18,000 wells listing spud dates, about 70% were drilled prior to 1980. Wells drilled before 1980 have a higher risk of well casing failures and are more likely to be sources of groundwater contamination.

Additionally, wells plugged prior to 1953 are not considered effective, even by industry standards. Prior to 1950, wells either were orphaned or plugged and abandoned with very little cement. Plugging was focused on protecting the oil reservoirs from rain infiltration rather than to “confine oil, gas and water in the strata in which they are found and prevent them from escaping into other strata.” Of the wells with drilling dates in the regulatory data, 30% are listed as having been drilled prior to the use of cement in well plugging.

With a total of over 245,000 wells in the state database, and considering the lack of monitoring prior to 1950, it’s reasonable to assume there are over 80,000 improperly plugged and unplugged wells in California.

Map of California’s Plugged Wells

View map fullscreen | How FracTracker maps work

The regions with the highest counts of plugged wells are the Central Valley and the South Coast. The top 10 county ranks are listed below in Table 1. Kern County has more than half of the total plugged wells in the entire state.

Table 1. Ranks of Counties by Plugged Well Counts
  • Rank
  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • County
  • Kern
  • Los Angeles
  • Orange
  • Fresno
  • Ventura
  • Santa Barbara
  • Monterey
  • San Luis Obispo
  • Solano
  • Yolo
  • Plugged Well Count
  • 65,733
  • 17,139
  • 7,259
  • 6,970
  • 4,302
  • 4,192
  • 2,266
  • 1,463
  • 1,456
  • 1,383

The issue is not unique to California. Nationally, an estimated 2.56 million oil and gas wells have been drilled and 1.93 million wells had been abandoned by 1975. Using interpolated data, the EPA estimates that as of 2016 there were 3.12 million abandoned wells in the U.S. and 69% of them were left unplugged.

In 2017, FracTracker Alliance organized an exercise to track down the locations of Pennsylvania’s abandoned wells that are not included in the PA Department of Environmental Protection’s digital records. Using paper maps and the FracTracker Mobile App, volunteers explored Pennsylvania woodlands in search of these hidden greenhouse gas emitters.

What are the risks?

Emissions

Studies by Kang et al. 2014, Kang et al 2016, Boothroyd et al 2016, and Townsend-Small et al. 2016 have all measured methane emissions from abandoned wells. Both properly plugged and improperly abandoned wells have been shown to leak methane and other VOCs to the atmosphere as well as into the surrounding groundwater, soil, and surface waters. Leaks were shown to begin just 10 years after operators plugged the wells.

Well density

The high density of aging and improperly plugged wells is a major risk factor for the current and future development of California’s oil and gas fields. When fields with old wells are reworked using new technology, such as hydraulic fracturing, CO2 flooding, or solvent flooding (including acidizing, water flooding, or steam flooding), the injection of additional fluid and gas increases pressure in a reservoir. Poorly plugged or aging wells often lack the integrity to avoid a blowout (the uncontrolled release of oil and/or gas from a well). There is a consistent risk that formation fluids will be forced to migrate up the plugged wellbores and bypass the existing plugs.

Groundwater

In a 2014 report, the U.S. Geological Service warned the California State Water Resources Control Board that the integrity of abandoned wells is a serious threat to groundwater sources, stating, “Even a small percentage of compromised well bores could correspond to a large number of transport pathways.”

The California Council on Science and Technology (CCST) has also suggested the need for additional research on existing aquifer contamination. In 2014, they called for widespread testing of groundwater near oil and gas fields, which has still not occurred.

Leaks

In addition to the contamination of underground sources of drinking water, abandoned well failures can even create a pathway for methane and fluids to escape to Earth’s surface. In many cases, such as in Pennsylvania, Texas, and California, where drilling began prior to the turn of the 20th century, many wells have been left unplugged. Of the abandoned wells that were plugged, the plugging process was much less adequate than it is today.

If plugged wells are allowed to leak, surface expressions can form. These leaks can travel to the Earth’s crust where oil, gas, and formation waters saturate the topsoil. A construction supervisor for Chevron named David Taylor was killed by such an event in the Midway-Sunset oil field near Bakersfield, CA. According to the LA Times, Chevron had been trying to control the pressure at the well-site. The company had stopped injections near the well, but neighboring operators continued high-pressure injections into the pool. As a result, migration pathways along old wells allowed formation fluids to saturate the Earth just under the well-site. Tragically, Taylor fell into a 10-foot diameter crater of 190° fluid and hydrogen sulfide.

California regulations

Following David Taylor’s death in 2011, California regulators vowed to make urgent reforms to the management of underground injection, and new rules finally went into effect on April 1, 2018. These regulations require more consistent monitoring of pressure and set maximum pressure standards. While this will help with the management of enhanced oil recovery operations, such as steam and water flooding and wastewater disposal, the issue of abandoned wells is not being addressed.

New requirements incentivizing operators to plug and abandon idle wells will help to reduce the number of orphan wells left to the state, but nothing has been done or is proposed to manage the risk of existing orphaned wells.

Conclusion

Why would the state of California allow new oil and gas drilling when the industry refuses to address the existing messes? Why are these messes the responsibility of private landholders and the state when operators declare bankruptcy?

New bonding rules in some states have incentivized larger operators to plug their own wells, but old low-producing or idle wells are often sold off to smaller operators or shell (not Shell) companies prior to plugging. This practice has been the main source of orphaned wells. And regardless of whether wells are plugged or not, research shows that even plugged wells release fugitive emissions that increase with the age of the plug.

If the fossil fuel industry were to plug the existing 1.666 million currently active wells, there would be nearly 5 million plugged wells that require regular inspections, maintenance, and for the majority, re-plugging, to prevent the flow of greenhouse gases. This is already unattainable, and drilling more wells adds to this climate disaster.

By Kyle Ferrar, Western Program Coordinator, FracTracker Alliance

Ted Auch, PhD

Wicked Witch of the Waste

/by

The Great Plains has become the unconventional oil & gas industry’s dumping ground, prompting questions about the security and resilience of the bread basket and the underlying Ogalalla Aquifer

Back in December of 2016, FracTracker analyzed the growing link between injection wells that dispose fracking waste and “induced seismicity” [1], or human-caused earthquakes. Our compiled maps from this analysis (including Figure 1 below) show seismic activity in Kansas and Oklahoma along with Class II injection well volumes up through 2015. 

Figure 1. Earthquakes and Class II Injection Well Activity at the Kansas-Oklahoma Border

This link was given acute attention at that time as a result of the magnitude 5.8 earthquake in Pawnee, Oklahoma on September 3rd, 2016, followed closely by a 4.5 earthquake on November 1st.  The industry’s increased production of waste came home to roost 5 days later when a magnitude 5.0 quake struck a mile west of the “Cushing Hub,” the largest commercial crude oil storage center in North America. The Cushing Hub is capable of storing 54 million barrels of crude – the equivalent of 2.8 times the U.S. daily oil refinery capacity and 3.1 times the daily oil refinery capacity of all of North America.

Sunflower State of Affairs

Since we published this analysis and associated maps, Class II injection wells have been in the news several times across the Great Plains. An investigation by KSN News found that the Kansas Corporation Commission (KCC) improperly permitted over 2,000 Class II injection wells. The KCC stated that public comment periods for well proposals lasted just 15 days, instead of the correct number of 30 days. This amounts to 42% and 28% of the state’s active and total inventory of oil and gas waste receiving wells approved with inaccurate public notices.


Quail Oil & Gas LC’s Class II Salt Water Disposal (SWD) well, Morris County,
KS near Diamond Creek (Photo Courtesy of Karla jo Grimmett at South 500 photography)

According to Cindy Hoedel, a freelance journalist in Kansas, “the KCC responded to the investigation findings… by ruling that no remedy was needed and closing the docket.”

Attorneys representing the Sierra Club maintain that improper permitting by the KCC continued into the Fall of 2018:

“The significance is they are choking us off in terms of giving us less and less time to try to mount a protest, to submit any kind of comment, and that’s a lot,” Cindy Hoedel, a Matfield Green resident who has complained about earthquakes in her area, said… “These notices get published in these tiny little newspapers, and sometimes it might take us 15 days before we find it”

As Ms. Hoedel wrote in an email when I asked her to comment on issues relating to Kansas’ Class II injection wells:

“The Republican controlled Kansas Legislature is trying to fend off several proposed bills that would reform the KCC (the regulatory body that oversees the permitting of Class II underground injection control wells). Citizen challenges of individual applications for disposal and EOR [enhanced oil recovery] wells continue, with the KCC moving more aggressively than in the past to dismiss protestants before a hearing is held. Some of these dismissals are being challenged in appellate court. The activists’ view is that EPA, the SWDA [Safe Water Drinking Act] and Congress clearly intend for the public to be able to participate in the regulatory process; instead, KCC has written regulations that are effectively barriers to participation… Activists have questions about the large number of EOR wells being applied for in Kansas and what their true purpose is, given the insignificant amounts of oil being produced compared to high volumes of injected fluids. Another concern is that the injection well earthquakes in Oklahoma and Kansas continue, yet KCC refuses to add regs that would address seismic risk in permit applications. There is also a problem with harassment of citizens exercising their right to protest – Scott Yeargain and I were both turned in to the Kansas AG’s office by a KCC staffer on the bogus claim that we were practicing law without a license because we helped explain the convoluted process to other protesters.”

 

Grapes of Wrath

Meanwhile, across the border, Oklahoma City and its surrounding suburbs have become the San Francisco of the Great Plains, with regular earthquake swarms (including many that exceed magnitude 4.0). According to Think Progress reporter Samantha Page, despite the damages and lawsuits caused by these earthquakes, “for years, the state was slow to respond, while Gov. Mary Fallin (R) and others questioned the link to human activity.” 

Eventually, by the end of 2016, the Oklahoma Corporation Commission responded by implementing a ‘traffic light’ protocol, in which operations are paused or stopped altogether following earthquakes of certain magnitudes. For a time, the EPA demanded a moratorium on disposal across Class II wells injecting into the Arbuckle formation in “high seismically active focus areas.”

Chad Warmington, president of the Oklahoma Oil and Gas Association, said that this response by the EPA is “a stellar example of the inefficiency of the federal government…It’s akin to a newspaper telling us today the football scores from games played 15 months ago.”

In reporting on the industry’s response, journalist Paul Monies, buried the lead when he pointed out the following in his second to last paragraph:

“Wastewater recycling remains an expensive option compared to the low costs of disposal wells in Oklahoma. While operators can inject wastewater into formations other than the Arbuckle, Hatfield said other formations don’t accept water as easily and are at shallower depths.”

The Map

Our second stab at mapping the scale and scope of Class II injection wells across the Great Plains is slightly different than our first effort in a few ways:

  1. This iteration includes Class II Salt Water Disposal (SWD) Injection Wells in Nebraska, Oklahoma, and Kansas on one map. Clicking on a well reveals its location, well name, operator, and the volume of wastewater disposed. Volumes are presented annually for Nebraska and monthly for 2011 to 2017 for Oklahoma and Kansas. We also present annual sums for Oklahoma from 2006 to 2010.
  2. The map shows Arkansas and Platte River Basin boundaries, which contain the entire inventory of OK, NE, and KS Class II wells.
  3. We’ve included Hydrologic Unit Codes, which when zoomed in to the map, identify sub-watersheds, and the Ogalalla Aquifer boundary, courtesy of the USGS’s Sharon Qi.
  4. Finally, we’ve includedUS Forest Service Robert G. Bailey’s Ecoregions to give a sense for the types of ecosystems threatened by the O&G industry’s demand for suitable waste disposal sites

View Map Full Screen | To view the legend on this map, click the “layers” icon on the top left of the screen

Table 1, below, breaks down the volumes of oil and gas wastewater disposed in Oklahoma, Kansas and Nebraska. Volumes are measured in million barrels, with one barrel equivalent to 42 gallons. The number of Class II SWD (salt water disposal) injection wells in these states is separated to show the total number of wells permitted verse the number of wells that were active (receiving waste).

Table 1. Class II injection well volumes in 2017

In total, 3,385,700,000 barrels of wastewater were disposed in 5,975 injection wells in these three states in 2017. The volume of wastewater disposed has increased in recent years (Table 2).

Table 2. Cumulative Class II injection well volumes to 2017, annual percent changes, and likely 2018 and 2027 volumes

In Table 2, the theoretical annual volumes for 2018 and 2027 are predictions based on the average of linear, exponential, and polynomial models.

The Kansas-Oklahoma Border

It is critical that we analyze the Great Plains fracking waste ecosystem across state lines. There are several reasons for this, including the proximity of Kansas’ most active Class II wells to the Oklahoma border (Figure 2) and the potential for the KCC to use enhanced oil recovery wells in Kansas to dispose of Oklahoma’s fracking waste.

Figure 2. Class II injection well volumes for 2017 along the Kansas-Oklahoma border.

Collaboration between front line communities, non-profits like FracTracker Alliance, and groups like the Kansas Water Advocacy Team (WAT) will be crucial to understanding the impacts of waste disposal writ large.  It seems like the “food vs energy” nexus has come to a head in the heart of the U.S. Bread Basket. We’ll continue to highlight and map the issues associated with this topic in the coming months and years.

Data Download Links

The following links contain the data used in the above tables and map, for use in excel and with Geographic Information Systems (GIS).

Oklahoma Class II Salt Water Disposal Waste Volumes, operator, well name, API number, etc.
2006 to 2017 for 10,927 wells.
Kansas Class II Salt Water Disposal Waste Volumes, operator, well name, API number, etc.
2011 to 2017 for 7,038 wells.
Nebraska Class II Salt Water Disposal Waste Volumes, operator, well name, API number, etc.
1982 to 2017 for 173 wells

[1] To learn more about Induced Seismicity, read an exclusive FracTracker two-part series from former researcher with Virginia Tech Department of Geosciences, Ariel Conn: Part I and Part II.

Additionally, the USGS has created an Induced Earthquakes landing page as part of their Earthquake Hazards Program.

Karen Edelstein

The Growing Web of Oil and Gas Pipelines

/by

Although the vast majority of scientists agree that we must rapidly move away from fossil fuels to avoid a human-caused climate catastrophe by the end of this century, pipeline construction remains a big business.

Pipelines are the backbone of domestic fossil fuel use and for delivering fuels to terminals for international export. Yet aside from a few high-profile pipeline controversies that show up in the media, few Americans are aware of the vast network of pipelines that transport oil and gas products from sources of extraction to industry and end-use consumers.

The United States is crisscrossed by over 1.63 million miles of fossil fuel pipelines. This includes:

Many of the country’s pipelines have been built within the last few decades, and in recent years, construction of more has been spurred on by the fracking boom. The total mile count of crude oil pipelines (currently 79,000) has increased over 60% between 2004 and 2017.  Natural gas distribution and estimated service pipeline miles increased 72% between 1984 and 2017 (Figure 1).

Figure 1. Miles of natural gas distribution (1,296,157 miles) and estimated service (
927,052 miles) pipelines in the U.S., 1984-2017

Although total mileage for transmission pipelines slightly dropped between 2004 and 2017 (according to the Pipeline and Hazardous Materials Safety Administration), total mileage for Hazardous Liquids pipelines jumped 33% during that same period (Figures 2 and 3).

Figure 2 (above). Total miles of Hazardous Liquid pipelines in the U.S., 2004-2017
Figure 3 (below). Break down of Hazardous Liquid pipeline miles in the U.S by what they’re transporting, 2004-2017

Exporting natural gas

When natural gas is imported or exported, it’s transported in a liquefied form. The product occupies much less space as a liquefied natural gas (LNG) than it does in its gaseous form, making it easier to transport.

For many years, the United States was an importer of natural gas, until 2007, when this trend quickly reversed, coinciding with the “fracking boom” in the Marcellus Shale, as well as several other shale plays in Texas, Wyoming, and elsewhere.

Figure 4. U.S. imports of natural gas, which is transported as liquefied natural gas (LNG)

LNG facilities store and process natural gas to help move it between markets. Between 2010 and 2017, the number of LNG facilities increased from 122 to 152 (includes LNG storage facilities). This nearly 25% increase reflects the surplus of natural gas in the lower 48 states.

The U.S. began exporting LNG in 2016, especially to Europe and China, where demand is high. According to the United States Energy Information Administration (EIA), LNG exports doubled between 2016 and 2017 (Figure 5).

Figure 5. U.S. LNG exports between January, 2016 and October, 2017, are shown in the blue bars

Exports are again expected to double over 2018 levels by the end of 2019, reaching a storage capacity of 9.6 billion cubic feet per day. The US is now the third largest exporter of LNG, after Australia and Qatar.

The breakdown of LNG terminals —existing and future— according to FERC is shown below. These terminals receive LNG imports or ship out LNG for export. The shift from LNG import to export activity over time is quite striking. No new import facilities are currently in the planning phase, yet there are 19 export facilities proposed and another 10 already approved.  

Table 1. Import and Export LNG Terminals in the US: Current, Approved, and Proposed.

  Import Export
Current 12: Everett, MA; Cove Point, MD; Elba Island, GA; Lake Charles, LA; offshore Boston, MA (2); Freeport, TX; Sabine, LA; Hackberry, LA; Sabine Pass, LA; Pascagoula, MS; Peñuelas, PR) 3: (Cove Point, MD; Sabine, LA; Kenai, AK)
Approved 3: Corpus Christi, TX; Gulf of Mexico (2) 10: Hackberry, LA (2); Freeport, TX; Corpus Christi, TX; Sabine Pass, LA (2); Elba Island, GA; Lake Charles, LA (2); Gulf of Mexico
Proposed None 19: Pascagoula, MS;  Cameron Parish, LA (2); Brownsville, TX (3); Port Arthur, TX; Jacksonville, FL; Plaquemines Parish, LA (2); Calcasieu Parish, LA; Nikiski, AK; Freeport, TX; Coos Bay, OR; Corpus Christi, TX; La Fourche Parish, LA; Sabine Pass, LA; Galveston Bay, TX

The challenge of keeping up

One of the challenges in working on oil and gas-related environmental advocacy is that from week to week, there are always changes in pipeline status. New pipelines are announced, others are delayed, others are postponed, and in some cases, projects are cancelled or defeated. Pipelines that have been under construction for years go on line. Listings are piece-meal, sometimes very vague, and sometimes reported by third and fourth party sources.

FracTracker is committed to sorting through this information, and providing a window into the expansion of oil and gas infrastructure. We have mapped and assembled information on over 60,000 miles of new and proposed oil and gas transmission pipelines and mapped over 250 projects since 2017.

Of these 60,000 pipeline miles, almost 9,800 have been completed and/or are operating. Close to 7,500 miles were cancelled or defeated. This leaves another 42,700 miles of pipeline that are currently in the replacement, reversal, planning or construction stages. 

In the interactive map below, against a background of existing pipelines, we show the newest pipelines that have come “on the radar” since 2017. In addition we show LNG terminals, one of the main destinations for the gas that flows through the pipelines to the export market.

Updated U.S. pipeline and LNG terminal map

View Map Full Screen | How Our Maps Work

Our mapping process

FracTracker is dedicated to bringing transparency to the landscape of oil and gas development. We use mapping tools such as GIS (geographic information systems) to illuminate developments in oil and gas infrastructure expansion.

Where do we get our data?

We draw our information from new listings by the United States Energy Information Administration (EIA) and Sierra Club for natural gas projects. In addition, we find announcements about new crude oil and gas pipeline projects on RBN Energy’s website. 

After we create a composite list of pipelines, the research begins. We search the internet for references to each pipeline, looking for industry announcements, descriptions, news articles, and, most importantly, the docket listings of the Federal Energy Regulatory Commission (FERC).

FERC may release detailed maps of pipeline routes from the company’s Environmental Impact Statement (EIS), filed after operators have progressed past the initial phases of planning. On occasion, we’ll stumble across links to Google Earth files that grassroots groups have ground-truthed. We can convert these .kml files into our ArcGIS mapping software directly.

Digital cartography

How do we go from online pictures of maps to data that we can use in our interactive maps? For the most part, we use a process called georeferencing, also known in some circles as “rubber-sheeting”. One of the beauties of digital cartography and GIS is that through the magic of computing, we can add information about location to mapped information. This allows us to add different features to a map, such as roads or rivers, and ensure that they line up correctly.

Let’s say I have a .jpg (image) file of a pipeline map that crosses four counties in Indiana. The .jpg shows both the pipeline and the county boundaries. I can open my GIS program and add a reference basemap of the United States, which is similar to what you see when you open Google Maps. I can zoom in to Indiana and add a second GIS layer of Indiana’s counties (already built with coordinates in the digital information), and voila! It drops right into where Indiana is on my base map. Can I do this with the pipeline .jpg? Not yet!

I have to use the clues on the pipeline image to place it in the correct location on the GIS map. Luckily, my pipeline map has county boundaries on it, so I can line up the corners (or other shapes) on the pipeline image to where they are on my map that is “smart” about location using ground control points.

Once I’m satisfied that the map I’ve added is in the correct location, I carefully trace the path of the pipeline, saving it as a GIS layer. Because it’s drawn with its own location data included, it will always appear in future maps in the same place relative to the rest of Indiana.

That’s our process in a nutshell.

Want to see this done as a demo? Here’s a nice 10-minute YouTube video:

https://www.youtube.com/watch?v=PHtxbpboDro

By Karen Edelstein, Eastern Program Coordinator

destroyed home following pipeline explosion in San Bruno, CA
Guest Author

Unnatural Disasters

/by

Guest blog by Meryl Compton, policy associate with Frontier Group

Roughly half of the homes in America use gas for providing heat, hot water or powering appliances. If you use gas in your home, you know that leaks are bad – they waste money, they pollute the air, and, if exposed to a spark, they could spell disaster.

Our homes, however, are only the end point of a vast production and transportation system that brings gas through a network of pipelines all the way from the wellhead to our kitchens. There are opportunities for wasteful and often dangerous leaks all along the way – leaks that threaten the public’s health and safety and contribute to climate change.

How frequent are gas leaks?

Between January 2010 and November 2018, there were a reported 1,888 incidents that involved a serious injury, fatality or major financial loss related to gas leaks in the production, transmission and distribution system, according to data from the Pipeline and Hazardous Materials Safety Administration. These incidents caused 86 deaths, 487 injuries and over $1 billion in costs.

When gas lines leak, rupture, or are otherwise damaged, the gas released can explode, sometimes right in our own backyards. Roughly one in seven of the incidents referenced above – 260 in total – involved an explosion.

In September 2018, for example, a series of explosions in three Massachusetts communities caused one death, numerous injuries and the destruction of as many as 80 homes. And there are many more stories like it from communities across the U.S. From the 2010 pipeline rupture and explosion in San Bruno, California, that killed eight people and destroyed almost 40 homes to the 2014 disaster in New York City that destroyed two five-story buildings and killed eight people, these events serve as a powerful reminder of the danger posed by gas.

The financial and environmental costs

Gas leaks are also a sheer waste of resources. While some gas is released deliberately in the gas production process, large amounts are released unintentionally due to malfunctioning equipment, corrosion and natural causes like flooding. The U.S. Energy Information Administration estimates that 123,692 million cubic feet of gas were lost in 2017 alone, enough to power over 1 million homes for an entire year. That amount is likely an underestimate. On top of the major leaks reported to the government agency in charge of pipeline safety, many of our cities’ aging gas systems are riddled with smaller leaks, making it tricky to quantify just how much gas is lost from leaks in our nation’s gas system.

Leaks also threaten the stability of our climate because they release large amounts of methane, the main component of gas and a potent greenhouse gas. Gas is not the “cleaner” alternative to coal that the industry often makes it out to be. The amount of methane released during production and distribution is enough to reduce or even negate its greenhouse gas advantage over coal. The total estimated methane emissions from U.S. gas systems have roughly the same global warming impact over a 20-year period as all the carbon dioxide emissions from U.S. coal plants in 2015 – and methane emissions are likely higher than this amount, which is self-reported by the industry.

In most states, there is no strong incentive for gas companies to reduce the amount of leaked gas because they can still charge customers for it through “purchased gas adjustment clauses.” These costs to consumers are far from trivial. Between 2001 and 2011, Americans paid at least $20 billion for gas that never made it to their homes.

These and other dangers of gas leaks are described in a recent fact sheet by U.S. PIRG Education Fund and Frontier Group. At a time when climate change is focusing attention on our energy system, it is critical that communities understand the full range of problems with gas – including the ever-present risk of leaks in the extensive network of infrastructure that brings gas from the well to our homes.

Get the Fact Sheet

The alternative

We should not be using a fuel that endangers the public’s safety and threatens the stability of our climate. Luckily, we don’t have to. Switching to electric home heating and hot water systems and appliances powered by renewable energy would allow us to move toward eliminating carbon emissions from homes. Electric heat pumps are twice as efficient as gas systems in providing heat and hot water, making them a viable and commonsense replacement. Similarly, as the cost of wind and solar keep falling, they will continue to undercut gas prices in many regions.

It’s time to move beyond gas and create a cleaner, safer energy system.

By Meryl Compton, policy associate with Frontier Group, a non-profit think tank part of The Public Interest Network. She is based in Denver, Colorado.

Feature image at top of page shows San Bruno, California, following the 2010 pipeline explosion

Matt Kelso, BA

Getting Rid of All of that Waste – Increasing Use of Oil and Gas Injection Wells in Pennsylvania

/by

Oil and gas development generates a lot of liquid waste.

Some of the waste comes that comes out of a well is from the geologic layer where the oil and gas resources are located. These extremely saline brines may be described as “natural,” but that does not make them safe, as they contain dangerous levels of radiation, heavy metals, and other contaminants.

Additionally, a portion of the industrial fluid that was injected into the well to stimulate production, known as hydraulic fracturing fluid, returns to the surface.  Some of these substances are known carcinogens, while others remain entirely secret, even to the personnel in the field who are employed to use the additives.

The industry likes to remind residents that they have used this technique for more than six decades, which is true. What separates “conventional” fracking from developing unconventional formations such as the Marcellus Shale is really a matter of scale.  Conventional formations are often stimulated with around 10,000 gallons of fluid, while unconventional wells now average more than 10 million gallons per well.

Conventional Well: 10,000 gallons
Olympic Swimming Pool: 660,253 gal
Unconventional Well: >10 Million gal

In 2017 alone, Pennsylvania oil and gas wells generated 57,653,023 barrels (2.42 billion gallons) of liquid waste.

Managing the waste stream

Liquid waste can be reused to stimulate other oil and gas wells, but reuse concentrates the contaminant load in the fluid. There is a limit to this concentration that operators can use, even for this industrial purpose.

Another strategy is to decrease the volume of the waste through evaporation and other treatment methods. This also increases the contaminant concentration. Pennsylvania used to permit “treatment” of wastewater at sewage treatment facilities, before being forced to concede that the process was completely ineffective, and resulted in contaminating streams and rivers throughout the Commonwealth.

In many states, much of this waste is disposed of in facilities known as salt water disposal (SWD) wells, a specific type of injection well. These waste facilities fall under the auspices of the US Environmental Protection Agency’s Underground Injection Control (UIC) program. Such wells are co-managed with states’ oil and gas regulatory agencies, although the specifics vary by state.

Morrow Co. Ohio, Fishburn Walter II injection well
Morrow Co. Ohio, Fishburn Walter II injection well
Ashtabula OH, Brine Waste Truck
Ashtabula OH, Brine Waste Truck

These photos show SWD wells in other states, but what about in Pennsylvania?

The oil and gas industry in Pennsylvania has not used SWD wells as a primary disposal method, as the state’s geology has been considered unsuitable for this process.  For example, on page 67 of this 2009 industry report, the authors saw treatment of flowback fluid at municipal facilities as a viable option (before the process was  banned in 2011), but underground injection as less likely (emphasis added):

The disposal of flowback and produced water is an evolving process in the Appalachians. The volumes of water that are being produced as flowback water are likely to require a number of options for disposal that may include municipal or industrial water treatment facilities (primarily in Pennsylvania), Class II injection wells [SWDs], and on-site recycling for use in subsequent fracturing jobs. In most shale gas plays, underground injection has historically been preferred. In the Marcellus play, this option is expected to be limited, as there are few areas where suitable injection zones are available.

The ban on surface “treatment” being discharged into Pennsylvania waters has increased the pressure for finding new solutions for brine disposal.  This is compounded by the fact that the per-well volume of fluid injected into shale gas wells in the region has nearly tripled in that time period. Much of what is injected comes back up to the surface and is added to the liquid waste stream.

Chemically-similar brine from conventional wells has been spread on roadways for dust suppression. This practice was originally considered a “beneficial use” of the waste product, but the Pennsylvania Department of Environmental Protection (DEP) halted that practice in May 2018.

None of these waste management decisions make the geology in Pennsylvania suddenly suitable for underground injection, however, they do increase the pressure on the state to find a disposal solution.

Concerns with SWD wells

There are numerous concerns with salt water disposal wells.  In October 2018, the DEP held a hearing in Plum Borough, on the eastern edge of Allegheny County, where there is a proposal to convert the Sedat 3A conventional well to an injection well. Some of the concerns raised by residents include:

  • Fluid and/or gas migration- There are numerous routes for fluids and gas to migrate from the injection formation to drinking water aquifers or even surface water.  Potential conduits include coal mines, abandoned gas wells, water wells, and naturally occurring fissures in crumbling sedimentary formations.
  • Induced seismicity- SWD wells have been linked to increased earthquake activity, either by lubricating or putting pressure on old faults that had been dormant. Earthquakes can occur miles away from the injection location, and in sedimentary formations, not just igneous basement rock.
  • Noise, diesel pollution, loss of privacy, and road degradation caused by a constant stream of industrial waste haulers to the well location.
  • Complicating existing issues-  Plum Borough and surrounding communities are heavily undermined, and in fact the well bore goes right through the Renton Coal Mine (another part of which has been on fire for decades).  Mine subsidence is already a widespread issue in the region, and many fear that even small seismic events could exacerbate this.
  • Possibility of surface spill-  Oil and gas is, sadly, a sloppy industry, with unconventional operations having accumulated more than 13,000 violations in Pennsylvania since 2008.  If a major spill were to happen at this location, there is the possibility of release into Pucketa Creek, which drains into the Allegheny River, the source of drinking water for multiple communities.
  • Radioactivity and other contaminants- Flowback fluids are often highly radioactive, contain heavy metals, and other contaminants that are challenging to effectively clean.  The migration of radon gas into homes above the injection formation is also a possibility.

Download Sedat 3A File Review Documents

The current state of SWDs in Pennsylvania

Pennsylvania has numerous data sources for oil and gas, but they are not always in agreement. To account for this, we have mapped SWDs (and a five mile buffer around them) from two different data sources in the map below. The first source is a subset of SWD wells from a larger dataset of oil and gas locations from the DEP’s mapping website. The second source is from a Waste Facility Report, represented in pink triangles that are offset at an angle to allow users to see both datasets simultaneously in instances where they overlap.

Map of existing, proposed, and plugged salt water disposal (SWD) injection wells in Pennsylvania.

 View map fullscreen | How FracTracker maps work

According to the first data set of DEP’s oil and gas locations, Pennsylvania contains 13 SWDs with an active status, one SWD with a regulatory inactive status, and eight that are plugged. The Waste Facility Report shows 10 SWD wells total, including one well that was left out of the other data set in Annin Township, McKean County.

It is worth noting that Pennsylvania’s definition for an “active” well status is confusing, to put it charitably. It does not mean that a well is currently in operation, nor does it even mean that it is currently permitted for the activity, whether that is waste disposal or gas production, or some other function. An active status means that the well has been proposed for a given use, and the well hasn’t been plugged, or assigned some other status.

The Sedat 3A well in Plum, for example, has an active status, although the DEP has not yet granted it a permit to operate as a SWD well. Another  status type is “regulatory inactive,” which is given to a well that hasn’t been used for its stated purpose in 12 months, but may potentially have some future utility.

Karst, coal mines, and streams

While there are numerous factors worthy of consideration when siting SWD wells, this map focuses on three: the proximity of karst formations, coal mines and nearby streams that the state designates as either high quality or exceptional value.

Karst formations are unstable soluble rock formations like limestone deposits which are likely to contain numerous subsurface voids. These voids are concerning in this context. For one reason, there’s the possibility of contaminated fluids and gasses migrating into underground freshwater aquifers. Also, the voids are inherently structurally unstable, which could compound the impacts of artificially-induced seismic activity caused by fluid injections in the well.

Our analysis found over 78,000 acres (123 square miles) of karst geology within five miles of current, proposed, or plugged SWD wells in Pennsylvania.

Coal mines, while a very different sedimentary formation, have similar concerns because of subsurface voids. Mine subsidence is already a widespread problem in many of the communities surrounding SWD well sites.  Pennsylvania has several available data sets, including active underground mine permits and digitized mined areas, which are used in this map.  Active mine permits show current permitted operations, while digitized mine areas offer a highly detailed look at existing mines, including abandoned mines, although the layer is not complete for all regions of the state.

In Pennsylvania, there are 56,542 acres (88 square miles) of active mines within five miles of SWD wells. Our analysis found 97,902 acres (153 square miles) of digitized mined areas within five miles of SWD wells.  Combined, there are 139,840 acres (219 square miles) of existing and permitted mines within the 5 mile buffer zone around SWDs in Pennsylvania.

Streams with the designation “high quality” and “exceptional value” are the best streams Pennsylvania has to offer, in terms of recreation, fishing, and biological diversity. In this analysis, we have identified such streams within a five mile radius of SWD wells, irrespective of the given watershed of the well location.

While the rolling topography of Western Pennsylvania sheds rainwater in a complicated network of drainages, groundwater is not subject to that particular geography. Furthermore, groundwater regularly interacts with surface water through water wells, abandoned O&G wells, and natural seeps and springs. Therefore, it is possible for SWDs to contaminate these treasured streams, even if they are not located within the same watershed.

Altogether, there are 716 miles of high quality streams and 110 miles of exceptional value streams within 5 miles of the SWDs in this analysis.

Conclusion

For decades, geologists have concluded that the subsurface strata in Pennsylvania were not suitable for oil and gas liquid waste disposal in underground injection wells.  The fact that vast quantities of this waste are now being produced in Pennsylvania has not suddenly made it a suitable location for the practice.  If anything, additional shallow and deep wells have further fractured the sedimentary strata, thereby increasing the risk of contamination.

The only factor that has changed is the volume of waste being produced in the region. SWD wells in nearby Ohio and West Virginia have capacity issues from their own production wells, and it is not clear that the geologic formations across the border are that much better than in Pennsylvania. But as new wells are drilled and volumes of hydraulic fracturing fluid continue to spiral into the tens of millions of gallons per well, the pressure to open new SWD wells in the state will only increase.

Perhaps because of these pressures, DEP has become quite bullish on the technology:

Several successful disposal wells are operating in Pennsylvania and options for more sites are always being considered. The history of underground disposal shows that it is a practical, safe and effective method for disposing of fluids from oil and gas production.
Up against this attitude, residents are facing an uphill battle trying to prevent harm to their health and property from these industrial facilities in their communities.  Municipalities that have attempted to stand up for their residents have been sued by DEP to allow for these injection wells.  The Department’s actions, which put the interests of industry above the health of residents and the environment, is directly at odds with the agency’s mission statement:
The Department of Environmental Protection’s mission is to protect Pennsylvania’s air, land and water from pollution and to provide for the health and safety of its citizens through a cleaner environment. We will work as partners with individuals, organizations, governments and businesses to prevent pollution and restore our natural resources.
It’s time for DEP to live up to its promises.

By Matt Kelso, Manager of Data and Technology, FracTracker Alliance

Bird's eye view of an injection well (oil and gas waste disposal)
Ted Auch, PhD

A Disturbing Tale of Diminishing Returns in Ohio

/by

Utica oil and gas production, Class II injection well volumes, and lateral length trends from 2010-2018

The US Energy Information Administration (EIA) recently announced that Ohio’s recoverable shale gas reserves have magically increased by 11,076 billion cubic feet (BCF). This increase ranks the Buckeye State in the top 5 for changes in recoverable shale natural gas reserves between 2016 and 2017 (pages 31- 32 here). After reading the predictable and superficial media coverage, we thought it was time to revisit the data to ask a pertinent question: What is the fracking industry costing Ohio?

Recent Shale Gas Trends in Ohio

According to the EIA’s report, Ohio currently sits at #7 on their list of proven reserves. It is estimated there are 27,021 BCF of shale gas beneath the state (Figure 1).

Graph of natural gas reserves in different states 2016-2017

Figure 1. Proven and change in proven natural gas reserves from 2016 to 2017 for the top 11 states and the Gulf of Mexico (calculated from EIA’s “U.S. Crude Oil and Natural Gas Proved Reserves, Year-End 2017”).

There are a few variations in the way the oil and gas industry defines proven reserves:

…an estimated quantity of all hydrocarbons statistically defined as crude oil or natural gas, which geological and engineering data demonstrate with reasonable certainty to be recoverable in future years from known reservoirs under existing economic and operating conditions. Reservoirs are considered proven if economic producibility is supported by either actual production or conclusive formation testing. – The Organization of Petroleum Exporting Countries

… the quantity of natural resources that a company reasonably expects to extract from a given formation… Proven reserves are classified as having a 90% or greater likelihood of being present and economically viable for extraction in current conditions… Proven reserves also take into account the current technology being used for extraction, regional regulations and market conditions as part of the estimation process. For this reason, proven reserves can seemingly take unexpected leaps and drops. Depending on the regional disclosure regulations, extraction companies might only disclose proven reserves even though they will have estimates for probable and possible reserves. – Investopedia

What’s missing from this picture?

Neither of the definitions above address the large volume of water or wastewater infrastructure required to tap into “proven reserves.” While compiling data for unconventional wells and injection wells, we noticed that the high-volume hydraulic fracturing (HVHF) industry is at a concerning crossroads. In terms of “energy return on energy invested,” HVHF is requiring more and more resources to stay afloat.

OH quarterly Utica oil & gas production along with quarterly Class II injection well volumes:

The map below shows oil and gas production from Utica wells (the primary form of shale gas drilling in Ohio). It also shows the volume of wastewater disposed in Class II salt water disposal injection wells.


 View map fullscreen | How FracTracker maps work

Publications like the aforementioned EIA article and language out of Columbus highlight the nominal increases in fracking productivity. They greatly diminish, or more often than not ignore, how resource demand and waste production are also increasing. The data speak to a story of diminishing returns – an industry requiring more resources to keep up gross production while simultaneously driving net production off a cliff (Figure 2).

Graph of Utica permits in Ohio on a cumulative and monthly basis along with the average price of West Texas Intermediate (WTI) and Brent Crude oil per barrel from September, 2010 to December, 2018

Figure 2. Number of Utica permits in Ohio on a cumulative and monthly basis along with the average price of West Texas Intermediate (WTI) and Brent Crude oil per barrel from September 2010 to December 2018

The Great Decoupling of New Year’s 2013

In the following analysis, we look at the declining efficiency of the HVHF industry throughout Ohio. The data spans the end of 2010 to middle of 2018. We worked with Columbus-area volunteer Gary Allison to conduct this analysis; without Gary’s help this work and resulting map, would not have been possible.

A little more than five years ago today, a significant shift took place in Ohio, as the number of producing gas wells increased while oil well numbers leveled off. The industry’s permitting high-water mark came in June of 2014 with 101 Utica permits that month (a level the industry hasn’t come close to since). The current six-month permitting average is 25 per month.

As the ball dropped in Times Square ringing in 2014, in Ohio, a decoupling between oil and gas wells was underway and continues to this day. The number of wells coming online annually increased by 229 oil wells and 414 gas wells.

Graph showing Number of producing oil and gas wells in Ohio’s Utica Shale Basin from 2011 to Q2-2018

Figure 3. Number of producing oil and gas wells in Ohio’s Utica Shale Basin from 2011 to Q2-2018

Graph of Producing oil and gas wells as a percentage of permitted wells in Ohio’s Utica Shale Basin from 2011 to Q2-2018

Figure 4. Producing oil and gas wells as a percentage of permitted wells in Ohio’s Utica Shale Basin from 2011 to Q2-2018

Permits

The ringing in of 2014 also saw an increase in the number of producing wells as a percentage of those permitted. In 2014, the general philosophy was that the HVHF industry needed to permit roughly 5.5 oil wells or 7 gas wells to generate one producing well. Since 2014, however, this ratio has dropped to 2.2 for oil and 1.4 for gas well permits.

Put another way, the industry’s ability to avoid dry wells has increased by 13% for oil and 18% for gas per year. As of Q2-2018, viable oil wells stood at 44% of permitted wells while viable gas wells amounted to 71% of the permitted inventory (Figure 4).

Production declines

from the top-left to the bottom-right

To understand how quickly production is declining in Ohio, we compiled annual (2011-2012) and quarterly (Q1-2013 to Q2-2018) production data from 2,064 unconventional laterals.

First, we present average data for the nine oldest wells with respect to oil and gas production on a per day basis (Note: Two of the nine wells we examined, the Geatches MAH 3H and Hosey POR 6H-X laterals, only produced in 2011-2012 when data was collected on an annual basis preventing their incorporation into Figures 6 and 7 belwo). From an oil perspective, these nine wells exhibited 44% declines from year 1 to years 2-3 and 91% declines by 2018 (Figure 5). With respect to natural gas, these nine wells exhibited 34% declines from year 1 to years 2-3 and 79% declines by 2018 (Figure 5).

Figure 5. Average daily oil and gas production decline curves for the above seven hydraulically fractured laterals in Ohio’s Utica Shale Basin, 2011 to Q2-2018

Four of the nine wells demonstrated 71% declines by the second and third years and nearly 98% declines by by Q2-2018 (Figure 6). These declines lend credence to recent headlines like Fracking’s Secret Problem—Oil Wells Aren’t Producing as Much as Forecast in the January 2nd issue of The Wall Street Journal. Four of the nine wells demonstrated 49% declines by the second and third years and nearly 81% declines by Q2-2018 (Figure 7).

Figure 6. Oil production decline curves for seven hydraulically fractured laterals in Ohio’s Utica Shale Basin from 2011 to Q2-2018

Figure 7. Natural gas production decline curves for seven hydraulically fractured laterals in Ohio’s Utica Shale Basin from 2011 to Q2-2018

Fracking waste, lateral length, and water demand

from bottom-left to the top-right

An analysis of fracking’s environmental and economic impact is incomplete if it ignores waste production and disposal. In Ohio, there are 226 active Class II Salt Water Disposal (SWD) wells. Why so many?

  1. Ohio’s Class II well inventory serves as the primary receptacle for HVHF liquid waste for Pennsylvania, West Virginia, and Ohio.
  2. The Class II network is situated in a crescent shape around the state’s unconventional wells. This expands the geographic impact of HVHF to counties like Ashtabula, Trumbull, and Portage to the northeast and Washington, Athens, and Muskingum to the south (Figure 8).

Map of Ohio showing cumulative production of unconventional wells and waste disposal volume of injection wells

Figure 8. Ohio’s unconventional gas laterals and Class II salt water disposal injection wells. Weighted by cumulative production and waste disposal volumes to Q3-2018.

Disposal Rates

We graphed average per well (barrels) and cumulative (million barrels) disposal rates from Q3-2010 to Q3-2018 for these wells. The data shows an average increase of 24,822 barrels (+1.05 million gallons) per well, each year.

That’s a 51% per year increase (Figure 9).

A deeper dive into the data reveals that the top 20 most active Class II wells are accepting more waste than ever before: an astounding annual per well increase of 728,811 barrels (+30.61 million gallons) or a 230% per year increase (Figure 10). This divergence resulted in the top 20 wells disposing of 4.95 times the statewide average between Q3-2010 and Q2-2013. They disposed 13.82 times the statewide average as recently as Q3-2018 (Figure 11).

Cumulative and average well volumes for all 226 of Ohio’s Class II Salt Water Disposal wells from Q3-2010 to Q3-2018

All of this means that we are putting an increasing amount of pressure on fewer and fewer wells. The trickle out, down, and up of this dynamic will foist a myriad of environmental and economic costs to areas surrounding wells. As an example, the images below are injection wells currently under construction in Brookfield, Ohio, outside Warren and minutes from the Pennsylvania border.

Image of construction on injection wellSeneca Resources Class II Salt Water Disposal Injection well under construction, Brookfield, Ohio
Image of construction on injection well
Image of construction on injection well
road leading up to injection well
road leading up to injection well
Overhead view of injection well
Overhead view of injection well

More concerning is the fact that areas of Ohio that are injection well hotspots, like Warren, are proposing new fracking-friendly legislation. These disturbing bills would lubricate the wheels for continued expansion of fracking waste disposal and permitting. House bills 578 and 393 and Senate Bill 165 monetize and/or commodify fracking waste by giving townships a share of the revenue. Such bills “…would only incentivize communities to encourage more waste to come into their existing inventory of Class II… wells, creating yet another race to the bottom.” Co-sponsors of the bills include Democratic Reps. Michael O’Brien, Glenn Holms, John Patterson, and Craig Riefel.

Lateral Lengths

The above trends reflect an equally disturbing trend in lateral length. Ohio’s unconventional laterals are growing at a rate of 9.1 to 15.6%, depending on whether you buy that this trend is linear or exponential (Figure 12). This author believes the trend is exponential for the foreseeable future. Furthermore, it’s likely that “super laterals” in excess of 3-3.5 miles will have a profound impact on the trend. (See The Freshwater and Liquid Waste Impact of Unconventional Oil and Gas in Ohio and West Virginia.)

This lateral length increase substantially increases water demand per lateral. It also impacts Class II well disposal rates. The increase accounts for 76% of the former and 88% of the latter when graphed against each other (Figure 13).

Figure 12. Ohio Utica unconventional lateral length from Q3-2010 to Q4-2018

Figure 13. Ohio Utica unconventional water demand and Class II SWD injection well disposal volumes vs lateral length from Q3-2010 to Q4-2018.

Conclusion

This relationship between production, resource demand, and waste disposal rates should disturb policymakers, citizens, and the industry. One way to this problem is to more holistically price resource utilization (or stop oil and gas development entirely).

Unfortunately, states like Ohio are practically giving water away to the industry.

Politicians are constructing legislation that would unleash injection well expansion. This would allow disposal to proceed at rates that don’t address supply-side concerns. It’s startling that an industry and political landscape that puts such a premium on “market forces” is unwilling to address these trends with market mechanisms.

We will continue to monitor these trends and hope to spread these insights to states like Oklahoma and Texas in the future.

By Ted Auch, Great Lakes Program Coordinator, FracTracker Alliance – with invaluable data compilation assistance from Gary Allison

Data Downloads

FracTracker is a proponent of data transparency, and so we often share the data we use to construct our maps analyses. Click on the links below to download the data associated with the present analysis:

  • OH Utica laterals

    Ohio’s Utica HVHF laterals as of December 2018 in length (feet) (zip file)

  • Wastewater disposal volumes

    Inventory of volumes disposed on a quarterly basis from 2010 to Q3-2018 for all 223 active Class II Salt Water Disposal (SWD) Injection wells in Ohio (zip file)

Matt Kelso, BA

Pennsylvania Drilling Trends in 2018

/by

With the new year underway, it’s an opportune moment to reflect on the state of unconventional oil and gas extraction in Pennsylvania and examine a few of the drilling trends. A logical place to start is looking at the new wells drilled in 2018.

As always, but perhaps even more so than in previous years, unconventional drilling in Pennsylvania is a tale of two shale plays, with hotspots in the southwestern and northeastern corners of the state. The northeastern hotspot seems to be extending westward, including 25 new wells in Jones Township in Elk County (an area shown in dark red near the “St Marys” label on the map). In the southwestern hotspot, the industry continues to encircle Allegheny County, closing in on the City of Pittsburgh like a constrictor.

Screen shot showing spud report for Indiana Township, Allegheny County from 1/1/2017 through 1/4/2019. We suspect these spud dates of 11/29/17 and 11/30/17 are incorrect.

Screen shot showing spud report for Indiana Township, Allegheny County from 1/1/2017 through 1/4/2019. We suspect these spud dates of 11/29/17 and 11/30/17 are incorrect.

Data error? As Pittsburgh-area residents reflect on the past year, some of them must be wondering why a new well pad in Indiana Township, just northeast of the city isn’t shown on the map above. The answer is that the data the Department of Environmental Protection (DEP) has for these wells indicate they were drilled November 29-3o, 2017, although we believe this to be incorrect. FracTracker obtained the data from the Spud Report on January 2, 2019, which indicates seven wells spudded in that two day span on the “Miller Jr. 10602” well pad. This activity drew considerable opposition from families in the Fox Chapel School district in May of 2018, and was therefore widely reported on by the media. An article published on WESA indicates an expected drill date of July 2018, for example.

It turns out the new year is also a good time to remember that our understanding of the oil and gas industry around us is shaped, molded, and limited by the availability and quality of the data. We brought the Indiana Township data error to the attention of DEP, which only confirmed that the operator (Range Resources) entered the spud dates into the DEP’s online system. Perhaps these well were drilled in November of 2018 not 2017? There is even a possibility these wells have yet to be drilled.

Here are a few more dissections of the data, such as it is:

Graph of unconventional (fracking) wells drilled in PA, YTD - Drilling trends

Figure 1: Unconventional wells drilled in PA by year: 2005 to 2018

Wells Drilled Over Time

Barring more widespread data issues, the status of a handful of wells in Indiana Township does not have much of an impact on the overall trend of drilling in the state. There were 779 wells on the report, representing just under 40% of the total from the peak year of 2011, when industry drilled 1,958 wells. The year 2019 was the fourth year in a row where the industry failed to drill 1,000 wells, averaging 719 per year over that span. In contrast, the five years between 2010 and 2014 saw an average of 1,497 wells per year, more than twice the more recent average. As mentioned in our Hazy Future report, projections based on very aggressive drilling patterns are already proving to be out of phase with reality, although petrochemical commodity markets might change drastically in the coming decades.

How long before wells are plugged?

We also like to periodically check to see how long these wells stay in service. In Pennsylvania, there are two relevant well statuses worth following: plugged and regulatory inactive. While there are a number of conditions that characterize regulatory inactive wells, they are essentially drilled wells that are not currently in production, but may have “future utility.” Therefore, the wells are not required to be permanently plugged at this time.

Unconventional wells drilled since 2005 in PA - Drilling trends

Figure 2: This chart shows the percentage of unconventional wells drilled since 2005 with a plugged or regulatory inactive status as of December 31, 2018.

In order to understand some of the finer points, it’s best to use Figure 1 (above) in conjunction with Figure 2. We can see that most of the wells drilled in the initial years of the Marcellus boom have already been plugged, although Figure 1 shows us that the sample size is fairly low for these years. In 2005, for example, 7 of the 9 (78%) unconventional wells drilled in the state that year are already plugged. The following year, 24 of the 37 (65%) wells drilled are now plugged, and an additional 4 (11%) wells have a regulatory inactive status as of the end of 2018. The following year, the combined plugged and inactive wells account for just over 50% of the 113 wells drilled that year, and this trend continues along a fairly predictable curve. An exception is the noticeable bump around the most active drilling years of 2010 and 2011, where there are slightly more wells with a plugged or inactive status than might be expected. It is interesting to note that even the most recent wells are not immune to being plugged, including 8 plugged wells and 4 inactive wells drilled in 2018 that were not able to get past their very first year in production.

Overall, of the 11,675 drilled wells accounted for on this graphic, 851 (7%) are plugged already, with an additional 572 (5%) of wells with an inactive status.  Unconventional wells that are 11 years old have a roughly 50% chance of being plugged or inactive, and we would therefore expect to see the number of these wells skyrocket in the coming years before leveling off, roughly mirroring the drilling boom and subsequent slowdown of Marcellus Shale extraction in Pennsylvania.

Conclusions

Many factors contribute to fluctuations in drilling trends for the Marcellus Shale and other unconventional wells in Pennsylvania. Very cold winters result in high consumption by residential and commercial users. New gas-fired power plants can increase the demand for additional drilling. Recessions and economic conditions are known to reduce the demand for energy as well, and drillers’ heavy debt burdens can slow down operations appreciably. Additionally, other fossil fuel and renewable energy sources compete with one another, altering the market conditions even further. And finally, every oil and gas play eventually reaches a point where the expected results from new wells are not worth the money required to get the hydrocarbons to the surface, and unconventional wells are much more expensive to develop than more traditional operations.

Because of all of these variables, month to month or even year to year fluctuations are not necessarily that telling.  On the other hand, a four-year period where drilling is roughly half of previous extraction is significant, and can’t be easily dismissed as a blip in the data.

By Matt Kelso, Manager of Data and Technology, FracTracker Alliance

Re: Falcon ethane pipeline project
FracTracker Alliance

Falcon Pipeline Moves Forward Despite Unresolved Concerns

/by

Pittsburgh, PA – Yesterday, the Pennsylvania Department of Environmental Protection (DEP) announced their decision to issue a permit for the construction of Shell’s Falcon ethane pipeline project in southwest PA. FracTracker Alliance is extremely disappointed that DEP is allowing this project to proceed despite heavy opposition from the public and unaddressed concerns for the safety and well-being of nearby residents and the surrounding environment.

The past year has seen countless issues from the construction of new pipelines in the Commonwealth – from hundreds of “inadvertent returns,” (spills of bentonite drilling mud) along the path of the Mariner East II project to the catastrophic explosion of the week-old Revolution Pipeline in Beaver County. These reoccurring and serious incidents make it clear that oil and gas midstream companies are rushing to put infrastructure in place, and DEP and other regulatory agencies have been failing in their mission to adequately supervise the process.

According to data from the US Pipeline and Hazardous Materials Safety Administration, there were 108 pipeline incidents in Pennsylvania between January 2010 and mid-July 2018, resulting in 8 fatalities, 15 injuries, requiring over 1,100 people to be evacuated from their homes, and causing more than $66 million in property damage. This track record, which does not include the Revolution Pipeline explosion in September of 2018, is frankly unacceptable.

Certainly, the Commonwealth has invested heavily in the Shell Ethane Cracker facility, offering steep tax subsidies and even paying the global petrochemical giant $2.10 for every barrel of ethane it consumes from Pennsylvania wells, equivalent to $1.6 billion over the next 25 years. It appears to FracTracker that these business arrangements have made the continued extraction and exploitation of hydrocarbons the priority for DEP, not protecting the environment and health and safety of Pennsylvanians, as the mission of the Department suggests is their focus. DEP’s decision also traces an unfortunate pattern of opaqueness and poor timing by announcing unpopular decisions right before the holidays.

Fundamentally, oil and gas companies like Shell exist to make profits, and will therefore make decisions to maximize earnings and limit their costs, if left to their own devices. This approach is often directly at odds with public safety, so Pennsylvania entrusts DEP to oversee the operations. FracTracker feels that with their decision to move forward with the project on December 20, 2018, DEP brushed over dozens of substantial concerns regarding the Falcon ethane pipeline project, and therefore failed in this mission. We remain unconvinced that the “appropriate construction techniques and special conditions” required by DEP will adequately protect the environment and health and safety of residents along the Falcon pipeline route.

Dec. 21st Update: After this article was written, FracTracker learned that Ohio’s EPA issued an air quality permit for the cracker plant in Belmont County, Ohio on December 21st. The short public comment period and the rush to issue permits again illustrates that significant public health and environmental concerns are given minimal importance versus corporate wishes and political expediency. The regulatory paradigm is broken. The public has been ill served by the agencies entrusted to safeguard their interests. A collective regional voice should be raised in protest.

About FracTracker Alliance

Started in 2010 as a southwestern Pennsylvania area website, FracTracker Alliance is now a national organization with regional offices across the United States in Pennsylvania, Washington DC, New York, Ohio, and California. The organization’s mission is to study, map, and communicate the risks of oil and gas development to protect our planet and support the renewable energy transformation. Its goal is to support advocacy groups at the local, regional and national level, informing their actions to positively shape our nation’s energy future. www.fractracker.org

Learn more about FracTracker’s coverage of the Falcon ethane pipeline project by exploring the posts below:

Appalachia storage hub prospects map by FracTracker
Erica Jackson

Storing Natural Gas Liquids in Appalachia

/by

Last month, the Department of Energy (DOE) submitted a report titled Ethane Storage and Distribution Hub in the United States to Congress. The report sums up several other recent geologic studies and economic analyses that evaluate the potential to create a large petrochemical hub in southwest Pennsylvania, Ohio, West Virginia, and northeastern Kentucky.

Most people call this region Appalachia because of the mountains, or the Ohio River Valley because of the namesake river. The petrochemical industry looks deeper: they’ve branded it Shale Crescent USA, after the shale gas thousands of feet underground. This article summarizes recent developments on storing natural gas liquids, including ethane, in this region – whatever you prefer to call it.

Background

The United States currently produces more natural gas than any other country in the world, with much of the fracked gas coming from the Marcellus and Utica shales in Appalachia. The DOE report predicts that production in this region will continue growing from an estimated at 8.19 trillion cubic feet (Tcf) in 2017, to 13.55 Tcf in 2025 and 19.5 Tcf in 2050.

Natural Gas Production Estimates:

8.19 Tcf in 2017
13.55 Tcf in 2025
19.5 Tcf in 2050

In addition to oil and gas, fracking produces natural gas liquids (NGLs), such as ethane, propane, and butane. NGLs are a key component of the petrochemical industry, which takes these resources and converts them into plastics and resins. As industry extracts more natural gas, it will also be left with more NGLs to manage.

Hoping to profit off NGLs, the oil and gas industry is investing in petrochemical production. In the Appalachian basin, the DOE predicts that production of ethylene from ethane will reach 640,000 barrels a day by 2025 (this is 20 times the amount the region produced in 2013). The Gulf Coast of the U.S., as well as countries in Asia and the Middle East, are also growing their production capacities. Globally, ethylene production is projected to grow 31% from 2017 to 2025.

The rise of the petrochemical industry comes at a point when there’s an increasing global awareness of the disaster that is plastic pollution. As much as 12.7 million tons of plastic waste goes into the ocean each year, affecting over 700 species of marine animals. On land, plastic waste is often shipped to less developed nations, where it ends up polluting poor communities and contaminating their drinking water and air.

Nevertheless, politicians in PA, OH, and WV are working hard to attract petrochemical build-out in Appalachia. The region already houses much of the infrastructure needed for a petrochemical hub, such as fracked wells that pump out NGLs and processing plants to separate these liquids from the rest of the natural gas stream. One thing it’s missing, however, is significant capacity to store natural gas liquids – particularly ethane.

Why does industry need storage?

Ethane storage offers several benefits to the petrochemical industry. For one, it would serve as a steady supply of ethane for plants like ethane crackers, which “crack” ethane into ethylene to make polyethylene plastic. With this constant supply (transported to crackers via pipeline), plants can operate 24 hours a day, year round, and avoid using energy to shutdown and restart. Storage also allows industry to adapt to fluctuations in demand and price.

Another argument for expanding petrochemical activity in Appalachia is to diversify the industry’s geography. The current petrochemical hub in Texas and Louisiana (where over 95% of the country’s ethylene production takes place) is subject to extreme weather events. In 2017, Hurricane Harvey caused over half of the nation’s polyethylene production capacity to shut down. The report mentions “extreme weather events” multiple times as justification for building a petrochemical hub in Appalachia. This stance strongly suggests that the DOE is preparing for increased hurricanes and flooding from climate change, although this is never explicitly stated. Unsurprisingly, the industry’s role in causing climate change is left out from the report as well.

What does storage look like?

While the term ‘natural gas liquid’ may seem like an oxymoron, it refers to the different forms the substances take depending on temperature and pressure. At normal conditions, NGLs are a gas, but when pressurized or exposed to extremely cold temperatures,  they act as a liquid. NGLs occupy significantly less space as a liquid, and are therefore moved and stored as a pressurized or refrigerated liquid.

Storage can be in above ground tanks, but is often underground in gas fields or underground caverns. NGLs are highly volatile, and storing them above ground puts workers and surrounding communities at risk. For example – last week, an above ground storage tank exploded at a natural gas processing plant in Washington County, PA, sending four people to the hospital. While underground storage is perceived as “safer,” it still poses significant risks, particularly in a geography like Appalachia full of wells, coal mines, and pipelines. This underground infrastructure can cause NGLs to leak during storage or the land above them to collapse.

A study out of West Virginia University, titled “A Geologic Study to Determine the Potential to Create an Appalachian Storage Hub For Natural Gas Liquids,” identified three different types of storage opportunities along the Ohio and Kanawha river valleys:

Underground storage options

  1. Mined-rock cavern: Companies can mine caverns in formations of limestone, dolomite, or sandstone. The formation must be at least 40 feet thick to hold NGLs. This study focused on formations of the Greenbrier Limestone, which occurs throughout southwestern Pennsylvania, West Virginia, and Kentucky.
  2. Salt cavern: Developing salt caverns involves injecting water underground to create a void, and then pumping NGLs into the cavern. Suitable salt caverns have “walls” at least 100 feet thick above and below the cavern. The study recommended salt caverns 1,500 to 3,000 feet deep, but considered those as deep as 6,700 feet.
  3. Gas field: NGLs can also be stored in natural gas fields or depleted gas fields in underground sandstone reservoirs. Suitable gas fields are 2,000 feet deep or more according to the WVU study.

Where could storage sites be located?

The West Virginia University study identified and ranked thousands of gas fields, several salt caverns, and many regions in the Greenbrier Limestone that could serve as NGL storage. Most of the top-ranked opportunities are in West Virginia, near the state’s borders with Ohio and Pennsylvania, and several cross beneath the Ohio or Kanawha rivers. The researchers conclude with three “prospects,” which are circled in Figure 1.

A map of storing natural gas liquids opportunities in the Ohio River Valley

Figure 1. NGL storage opportunities identified by the Appalachian Oil and Natural Gas Consortium at West Virginia University

The table below lists the specific storage opportunities in each prospect, as well as the available data on depth, thickness, and acreage of the formations. Also listed are the counties that the storage facility would cross into.

Name Type Depth (feet) Thickness (feet) Counties Land Size (acres)
Salina F4 Salt cavern Salt cavern – >100 to 150 Primarily Columbiana, OH, also Hancock, WV & Beaver, PA 83,775
Salina F4 salt cavern Salt cavern – 100 to 150 Primarily Jefferson, OH, also Brooke & Hancock WV, & Washington, PA 129,017
Ravenna-Best Consolidated Field Depleted gas field 4,107 to 6,497 25 to 156 Mahoning, OH 69,000
No specific field was ranked Gas field in Oriskany sandstone 3,000 to 7,000 0 to 70+ Throughout the prospect

Existing NGL Storage

Storage in the United States

The U.S. has two major NGL storage hubs (both in salt caverns): One is in Mont Belvieu, Texas and the other in Conway, Kansas. These facilities are strategically located near the petrochemical industry’s hub along the Gulf Coast. There is also underground storage in Sarnia, Ontario.

Industry in Appalachia is connected to these storage facilities via pipelines, including Sunoco’s Mariner West that transports ethane to Sarnia, and the Appalachia-Texas-Express (ATEX) pipeline that takes ethane to Mont Belvieu. However, as suggested above, NGL storage in Appalachia is also under development.

Appalachia Storage & Trading Hub

Appalachia Development Group LLC is heading the development of the Appalachia Storage & Trading Hub initiative. The company has not announced the specific location for underground storage, but has been working hard to secure the funds  for this development.

In September of 2017, Appalachia Development Group submitted part 1 of a 2-part application for a $1.9 billion loan to the US DOE Loan Program Office. The DOE approved the application the following January, inviting the company to submit the second part, which is currently pending. This second part goes through the DOE’s Title XVII innovative clean energy projects loan program.

According to the DOE, this program “provides loan guarantees to accelerate the deployment of innovative clean energy technology.” Paradoxically, this means the DOE may give clean energy funds to the petrochemical industry, which is fueled by fossil fuels and does not provide energy but rather plastic and resins.

Steven Hedrick, the CEO of Appalachia Development Group, was part of a West Virginia trade delegation that traveled to China in 2017 to meet with China’s largest energy company. This meeting, which included President Trump and China’s President Xi Jinping, resulted in China Energy agreeing to invest $83.7 billion to support natural gas and petrochemical development in West Virginia. (Of note: This agreement has faced uncertainty following Trump’s tariffs on Chinese goods). West Virginia Governor Jim Justice later criticized Hedrick’s involvement in the meeting, where he promoted the interests of his private company.

Mountaineer NGL Storage Project

Another company, Energy Storage Ventures LLC, has plans to construct NGL storage near Clarington, Ohio. This facility would be on land formerly belonging to Quarto Mining Company’s Powhatan Mine No. 4. Called “Mountaineer NGL Storage,” the project would develop salt caverns to store propane, ethane, and butane. Each cavern could store 500,000 barrels (21 million gallons) of NGLs.

The video below, made by the Energy Storage Ventures, describes the process of developing salt caverns for storage.

The Mountaineer NGL Storage Project location is about 12 miles south of the PTTGC ethane cracker (if built), in Dilles Bottom Ohio. It’s also roughly 60 miles south of the Shell ethane cracker (under construction) in Potter Township, PA. If developed, the project could supply these plants with ethane and allow them to continuously operate. According to Energy Storage Ventures President, David Hooker, the project would also trigger $500 million in new pipelines in the region and $1 billion in fractionation facilities to separate NGLs.

Energy Storage Ventures wants to build three pipelines beneath the Ohio River. Two pipelines (one for ethane and one for propane and butane) would deliver NGLs to the storage site from Blue Racer Natrium, a fractionation plant that separates dry natural gas from NGLs. A third pipeline would take salt brine water from the caverns to the Marshall County chlorine plant (currently owned by Westlake Chemical Corp). These facilities, as well as the locations of the two ethane crackers storage could serve, are in the map below. This map also includes the potential storage opportunities the researchers at West Virginia University identified.

View map full screen | How FracTracker maps work

Referring to concerns about building pipelines and caverns near the Ohio River, a drinking water source for 5 million people, the company’s president David Hooker stated, “This is not rocket science. These things have operated safely for years… Salt, at depth, is impermeable. You won’t see any migration out of the salt.”

This video is a rendering of what the 200-acre site will look like, including the salt water impoundment structure (capable of holding 3.25 million barrels), and the infrastructure needed to deliver products and equipment by rail and truck:

The company has stated that it owns both the land and mineral rights it needs to develop the caverns, but the project has also faced delays.

Where is this plastic going?

One common argument for a petrochemical hub in Appalachia is the region’s proximity to the downstream sector of petrochemical industry. Manufacturers such as PPG Industries, Dow Chemical Inc., and BASF are all based in the area and could make use of the feedstock from an Appalachian hub.

However, the report doesn’t make it clear where the plastic and resin end products will land. It does state that the demand in the United States isn’t enough to swallow up two major petrochemical hubs worth of plastic.

Export markets

The DOE report states that, “the development of new petrochemical capacity in Appalachia is not necessarily in conflict with Gulf Coast expansion.” Since the Gulf Coast already has the infrastructure for export, it could focus on international markets while Appalachia meets domestic demand. Alternatively, the Appalachian hub could serve European destinations while the Gulf Coast hub delivers to Pacific Basin and South American destinations. Plastic consumption is highly correlated with population, so countries with large, growing populations such as India and China are likely markets.

It’s important to note that the U.S. isn’t the only country increasing its production of petrochemical derivatives, and as the report notes, exports from the US “may face a challenge from global capacity surplus.” Figure 2 shows that global production of ethylene is expected to surpass global consumption, shown in Figure 3. The graph of consumption likely ignores the impact of plastic-reducing policies that hundreds of countries and cities are implementing. As such, it may be an over-estimation.

Historical and Projected Ethylene Production Capacity by Global Area

Figure 2. Historical and future ethylene production by global region. Source

Graph of ethylene consumption by global area.

Figure 3. Ethylene consumption by global region. Source

In the end, it appears that the industry’s plan is to build first, and worry about markets later, hoping that a growing supply of affordable plastic will increase consumption.

Perhaps the reason industry is so eager to forge a market is because oil and gas is struggling with a lot of debt. A study out of the Sightline Institute found that as of the first half of 2018, “US fracking-focused oil and gas companies continued their eight-year cash flow losing streak.”  The Center for International Environmental Law found that petrochemicals generally have a larger profit margin than oil and gas: “In 2015, ExxonMobil’s Chemicals segment accounted for roughly 10% of its revenues but more than 25% of its overall profits.”

Plastic is one way to subsidize this dying industry…

Beyond Storing Natural Gas Liquids

The motive behind developing storage is to catalyze and support a major industry. The DOE report states that the new infrastructure required “would include gathering lines, processing plants, fractionation facilities, NGLs storage facilities, ethane crackers, and then…plants for polyethylene, ethylene dichloride, ethylene oxide, and other infrastructure.” A hub would require more fracking and wastewater injection wells, cause even more heavy truck traffic that adds stress to roadways, and require additional power plant capacity to serve its electricity demand.

In other words, an Appalachian petrochemical hub would profoundly impact the region. The report contains an in-depth analysis of the economic impacts, but fails to mention any environmental concerns, social impacts on communities, or health effects. The other major studies on this buildout,  mentioned above, follow a similar pattern.

A quick look at industry along the Gulf Coast tells you that environmental, social, and health concerns are very real and produce their own economic debts. The petrochemical industry has created a “cancer alley” in Texas and Louisiana, disproportionately impacting low-income and minority communities. Yet, industry is preparing another hub without a single comprehensive environmental impact assessment or health assessment for the region. As each pipeline, fracked well, and plant is permitted separately, we can’t properly assess the cumulative negative impacts this development will have on our waterways, forests, soil, or air quality. Therefore, we also won’t know how it will affect our health.

Looking into the future

The report analyzes the industry through 2050. It states that NGL output in Appalachia:

… will continue to grow throughout the forecast period. As natural gas production gradually migrates away from liquids-rich gas areas, which are expected to slowly deplete, to dryer areas, the rate of growth in NGPL production will slow relative to the rate of natural gas production growth.

In 31 years, the kids growing up in Appalachia right now could be left with brownfields, dried-up wells, and abandoned ethane crackers. But it doesn’t have to be this way. Last year, the DOE reported that there are more jobs in clean energy, energy efficiency, and alternative vehicles than in fossil fuels. By using funds such as the DOE’s Title XVII innovative clean energy loan – for actual clean energy – we can bring economic development to the region that will be relevant past 2050 and that won’t sacrifice our health and natural resources for short-term private gains.

By Erica Jackson, Community Outreach and Communications Specialist

Map of pipeline incidents across the US
Matt Kelso, BA

Pipeline Incidents Continue to Impact Residents

/by

Pipelines play a major role in the oil and gas extraction industry, allowing for the transport of hydrocarbons from well sites to a variety of infrastructure, including processing plants, petrochemical facilities, power generation plants, and ultimately consumers. There are more than 2.7 million miles of natural gas and hazardous liquid pipelines in the United States, or more than 11 times the distance from Earth to the moon.

With all of this infrastructure in place, pipelines are inevitably routed close to homes, schools, and other culturally or ecologically important locations. But how safe are pipelines, really? While they are typically buried underground and out of sight, many residents are concerned about the constant passage of volatile materials through these pipes in close proximity to these areas, with persistent but often unstated possibility that something might go wrong some day.

Safety talking points

In an attempt to assuage these fears, industry representatives and regulators tend to throw around variants of the word “safe” quite a bit:

Pipelines are the safest and most reliable means of transporting the nation’s energy products.
— Keith Coyle, Marcellus Shale Coalition

Although pipelines exist in all fifty states, most of us are unaware that this vast network even exists. This is due to the strong safety record of pipelines and the fact that most of them are located underground. Installing pipelines underground protects them from damage and helps protect our communities as well.
— Pipeline and Hazardous Materials Safety Administration (PHMSA)

Pipelines are an extremely safe way to transport energy across the country.
— Pipeline 101

Knowing how important pipelines are to everyday living is a big reason why we as pipeline operators strive to keep them safe. Pipelines themselves are one of the safest ways to transport energy with a barrel of crude oil or petroleum product reaching its destination safely by pipeline 99.999% of the time.
— American Petroleum Institute

But are pipelines really safe?

Given these talking points, the general public can be excused for being under the impression that pipelines are no big deal. However, PHMSA keeps records on pipeline incidents in the US, and the cumulative impact of these events is staggering. These incidents are broken into three separate reports:

  1. Gas Distribution (lines that take gas to residents and other consumers),
  2. Gas Transmission & Gathering (collectively bringing gas from well sites to processing facilities and distant markets), and
  3. Hazardous Liquids (including crude oil, refined petroleum products, and natural gas liquids).

Below in Table 1 is a summary of pipeline incident data from 2010 through mid-November of this year. Of note: Some details from recent events are still pending, and are therefore not yet reflected in these reports.

Table 1: Summary of pipeline incidents from 1/1/2010 through 11/14/2018

Report Incidents Injuries Fatalities Evacuees Fires Explosions Damages ($)
Gas Distribution 934 473 92 18,467 576 226 381,705,567
Gas Transmission & Gathering 1,069 99 24 8,614 121 51 1,107,988,837
Hazardous Liquids 3,509 24 10 2,471 111 14 2,606,014,109
Totals 5,512 596 126 29,552 808 291 4,095,708,513

Based on this data, on average each day in the US 1.7 pipeline incidents are reported (a number in line with our previous analyses), requiring 9 people to be evacuated, and causing almost $1.3 million in property damage. A pipeline catches fire every 4 days and results in an explosion every 11 days. These incidents result in an injury every 5 days, on average, and a fatality every 26 days.

Data shortcomings

While the PHMSA datasets are extremely thorough, they do have some limitations. Unfortunately, in some cases, these limitations tend to minimize our understanding of the true impacts. A notable recent example is a series of explosions and fires on September 13, 2018 in the towns of Lawrence, Andover, and North Andover, in the Merrimack Valley region of Massachusetts. Cumulatively, these incidents resulted in the death of a young man and the injuries to 25 other people. There were 60-80 structure fires, according to early reports, as gas distribution lines became over-pressurized.

The preliminary PHMSA report lists all of these Massachusetts fires as a single event, so it is counted as one fire and one explosion in Table 1. As of the November 14 download of the data, property damage has not been calculated, and is listed as $0. The number of evacuees in the report also stands at zero. This serves as a reminder that analysis of the oil and gas industry can only be as good as the available data, and relying on operators to accurately self-report the full extent of the impacts is a somewhat dubious practice.

View map fullscreen | How FracTracker maps work

This map shows pipeline incidents in the US from 1/1/2010 through 11/14/2018. Source: PHMSA. One record without coordinates was discarded, and 10 records had missing decimal points or negative (-) signs added to the longitude values. A few obvious errors remain, such as a 2012 incident near Winnipeg that should be in Texas, but we are not in a position to guess at the correct latitude and longitude values for each of the 5,512 incidents.

Another recent incident occurred in Center Township, a small community in Beaver County, Pennsylvania near Aliquippa on September 10, 2018. According to the PHMSA Gas Transmission & Gathering report, this incident on the brand new Revolution gathering line caused over $7 million in damage, destroying a house and multiple vehicles, and required 49 people to evacuate. The incident was indicated as a fire, but not an explosion. However, reporting by local media station WPXI quoted this description from a neighbor:

A major explosion, I thought it was a plane crash honestly. My wife and I jumped out of bed and it was just like a light. It looked like daylight. It was a ball of flame like I’ve never seen before.

From the standpoint of the data, this error is not particularly egregious. On the other hand, it does serve to falsely represent the overall safety of the system, at least if we consider explosions to be more hazardous than fires.

Big picture findings

Comparing the three reports against one another, we can see that the majority of incidents (64%) and damages (also 64%) are caused by hazardous liquids pipelines, even though the liquids account for less than 8% of the total mileage of the network. In all of the other categories, however, gas distribution lines account for more than half of the cumulative damage, including injuries (79%), deaths (73%), evacuees (62%), fires (71%), and explosions (78%). This is perhaps due to the vast network (more than 2.2 million miles) of gas distribution mains and service lines, as well as their nature of taking these hazardous products directly into populated areas. Comparatively, transmission and hazardous liquids lines ostensibly attempt to avoid those locations.

Is the age of the pipeline a factor in incidents?

Among the available attributes in the incident datasets is a field indicating the year the pipeline was installed. While this data point is not always completed, there is enough of a sample size to look for trends in the data. We determined the age of the pipe by subtracting the year the pipe was installed from the year of the incident, eliminating nonsensical values that were created when the pipeline age was not provided. In the following section, we will look at two tables for each of the three reports. The first table shows the cause of the failure compared to the average age, and the second breaks down results by the content that the pipe was carrying. We’ll also include a histogram of the pipe age, so we can get a sense of how representative the average age actually is within the sample.

A. Gas distribution

Each table shows some fluctuation in the average age of pipeline incidents depending on other variables, although the variation in the product contained in the pipe (Table 3) are minor, and may be due to relatively small sample sizes in some of the categories. When examining the nature of the failure in relation to the age of the pipe (Table 2), it does make sense that incidents involving corrosion would be more likely to afflict older pipelines, (although again, the number of incidents in this category is relatively small). On average, distribution pipeline incidents occur on pipes that are 33 years old.

When we look at the histogram (Figure 1) for the overall distribution of the age of the pipeline, we see that those in the first bin, representing routes under 10 years of age, are actually the most frequent. In fact, the overall trend, excepting those in the 40 t0 50 year old bin, is that the older the pipeline, the fewer the number of incidents. This may reflect the massive scale of pipeline construction in recent decades, or perhaps pipeline safety protocol has regressed over time.

Pipeline incidents charting

Figure 1. Age of pipeline histogram for gas distribution line incidents between 1/1/2010 and 11/14/2018. Incidents where the age of the pipe is unknown are excluded.

B. Gas Transmission & Gathering

Transmission & Gathering line incidents occur on pipelines routes that are, on average, five years older than their distribution counterparts. Corrosion, natural force damage, and material failures on pipes and welds occur on pipelines with an average age above the overall mean, while excavation and “other outside force” incidents tend to occur on newer pipes (Table 4). The latter category would include things like being struck by vehicles, damaged in wildfires, or vandalism. The contents of the pipe does not seem to have any significant correlation with the age of the pipe when we take sample size into consideration (Table 5).

The histogram (Figure 2) for the age of pipes on transmission & gathering line incidents below shows a more normal distribution, with the noticeable exception of the first bin (0 to 10 years old) ranking second in frequency to the fifth bin (40 to 50 years old).

It is worth mentioning that, “PHMSA estimates that only about 5% of gas gathering pipelines are currently subject to PHMSA pipeline safety regulations.” My correspondence with the agency verified that the remainder is not factored into their pipeline mileage or incident reports in any fashion. Therefore, we should not consider the PHMSA data to completely represent the extent of the gathering line network or incidents that occur on those routes.

Pipeline incidents chart

Figure 2. Age of pipeline histogram for transmission & gathering line incidents between 1/1/2010 and 11/14/2018. Incidents where the age of the pipe is unknown are excluded.

C. Hazardous Liquids

The average incident on hazardous liquid lines occurs on pipelines that are 27 years old, which is 6 years younger than for distribution incidents, and 11 years younger than their transmission & gathering counterparts. This appears to be heavily skewed by the equipment failure and incorrect operation categories, both of which occur on pipes averaging 15 years old, and both with substantial numbers of incidents. On the other hand, excavation damage, corrosion, and material/weld failures tend to occur on pipes that are at least 40 years old (Table 6).

In terms of content, pipelines carrying carbon dioxide happen on pipes that average just 11 years old, although there are not enough of these incidents to account for the overall departure from the other two datasets (Table 7).

The overall shape of the histogram (Figure 3) is similar to that of transmission & gathering line incidents, except that the first bin (0 to 10 years old) is by far the most frequent, with more than 3 and a half times as many incidents as the next closest bin (4o to 50 years old). Operators of new hazardous liquid routes are failing at an alarming rate. In descending order, these incidents are blamed on equipment failure (61%), incorrect operation (21%), and corrosion (7%), followed by smaller amounts in other categories. The data indicate that pipelines installed in previous decades were not subject to this degree of failure.

Pipeline incidents charting

Figure 3. Age of pipeline histogram for hazardous liquid line incidents between 1/1/2010 and 11/14/2018. Incidents where the age of the pipe is unknown are excluded.

Conclusions

When evaluating quotes, like those listed above, that portray pipelines as a safe way of transporting hydrocarbons, it’s worth taking a closer look at what they are saying.

Are pipelines the safest way of transporting our nation’s energy products? This presupposes that our energy must be met with liquid or gaseous fossil fuels. Certainly, crude shipments by rail and other modes of transport are also concerning, but movements of solar panels and wind turbines are far less risky.

Does the industry have the “strong safety record” that PHMSA proclaims? Here, we have to grapple with the fact that the word “safety” is inherently subjective, and the agency’s own data could certainly argue that the industry is falling short of reasonable safety benchmarks.

And what about the claim that barrels of oil or petroleum products reach their destination “99.999% of the time? First, it’s worth noting that this claim excludes gas pipelines, which account for 92% of the pipelines, even before considering that PHMSA only has records on about 5% of gas gathering lines in their pipeline mileage calculations. But more to the point, while a 99.999% success rate sounds fantastic, in this context, it isn’t good enough, as this means that one barrel in every 100,000 will spill.

For example, the Dakota Access Pipeline has a daily capacity of 470,000 barrels per day (bpd). In an average year, we can expect 1,715 barrels (72,030 gallons) to fail to reach its destination, and indeed, there are numerous spills reported in the course of routine operation on the route. The 590,000 bpd Keystone pipeline leaked 9,700 barrels (407,400 gallons) late last year in South Dakota, or what we might expect from four and a half years of normal operation, given the o.001% failure rate. In all, PHMSA’s hazardous liquid report lists 712,763 barrels (29.9 million gallons) were unintentionally released, while an additional 328,074 barrels (13.8 million gallons) were intentionally released in this time period. Of this, 284,887 barrels (12 million gallons) were recovered, meaning 755,950 barrels (31.7 million gallons) were not.

Beyond that, we must wonder whether the recent spate of pipeline incidents in new routes is a trend that can be corrected. Between the three reports, 1,283 out of the 3,853 (32%) incidents occurred in pipelines that were 10 years old or younger (where the year the pipeline’s age is known). A large number of these incidents are unforced errors, due to poor quality equipment or operator error.

One wonders why regulators are allowing such shoddy workmanship to repeatedly occur on their watch.

By Matt Kelso, Manager of Data and Technology, FracTracker Alliance