By Ryan Schnurr

Two hundred feet below the surface of the Straits of Mackinac—between Lakes Michigan and Huron, at the tip of the Michigan mitt—oil is moving through the Great Lakes. It is conducted by twin pipelines, 20 inches in diameter, which run parallel to each other across the bottom of the Straits, a thousand feet apart and supported by periodic joists. Together they represent 4.5 miles of Enbridge Line 5. Built in 1953, Line 5 is 645 miles long. It runs from Superior, Wisconsin, to Sarnia, Ontario. Outside of its twin pipelines in the Straits of Mackinac (pronounced “mack-in-aw”), Line 5 is a single pipeline, thirty inches across and buried in the ground. Around twenty million gallons of crude oil and natural gas liquids move through Line 5 every day.

Line 5 is part of a larger network of pipelines known as the Lakehead System. The Lakehead System is 5,022 miles long and operated by Enbridge Energy Partners, a division of the Canadian company Enbridge Inc. Collectively, its pipelines  deliver about 2.5 million barrels of crude oil every day to refineries on and around the lakes, and constitute Enbridge’s U.S. Mainline. It could be said at this point, thanks to a particularly sharp-edged homonym, that the Great Lakes region is mainlining oil.

Some of the oil in Line 5 originates thousands of miles to the northwest, in the oil sands of Alberta, Canada. Oil sands, or tar sands, are a combination of sand, clay, water, and bitumen—a heavy, viscous black oil. Bitumen shows up most commonly in asphalt and roofing tiles, but Canadian oil sands are often converted to a synthetic crude and sent the way of refineries. Before entering Line 5, bitumen is upgraded—the technical term—to a light synthetic crude.

“In really, really simple terms, there is nowhere worse—on earth—to have an oil sands pipeline system than the Great Lakes region.”

Some people think about the implications of this for a living. Dr. Rachel Havrelock is associate professor of English at the University of Illinois at Chicago and director of the UIC Freshwater Lab, where she oversees a collection of scholars and activists thinking about the future of the Great Lakes. She has written books and articles about the cultural and political dimensions of water and oil, especially in the Middle East, and is a member of the Petrocultures Research Group, based at the University of Alberta. In 2011, unrelated to pipelines, she wrote a hip-hop play, From Tel Aviv to Ramallah: A Beatbox Journey, which starred her husband, Yuri Lane, a professional beatboxer.

Dr. Havrelock’s office is on the nineteenth floor of UIC’s University Hall. Its windows face east. On a clear day, you can see Lake Michigan. On one wall of the office is a large map of Lakes Superior through Ontario. If you visit her office and start asking about the lakes, Dr. Havrelock will use this map as a reference. A couple of years ago, when I was a graduate student at UIC, I showed up with a question about a project I was working on and she went over to the wall and started pointing. Recently, I met up with her again and asked about the Lakehead System.

“In really, really simple terms, there is nowhere worse—on earth—to have an oil sands pipeline system than the Great Lakes region,” Dr. Havrelock said. She is a tall woman in her early forties with shaped brown hair and an enthusiastic face. She speaks deliberately and with authority. “It is, everything else aside, the world’s worst planning.”

The Great Lakes, collectively, are the largest available source of surface freshwater on the planet. Technically, the polar ice caps have more. Concentrated along the U.S.-Canada border from Minnesota to New York state, the Great Lakes are sometimes referred to as the “inland seas.” They contain twenty-one percent of the world’s fresh water supply and eighty-four percent of the fresh water in North America. Thirty-five million people on both sides of the border drink them.

“On every register—social, economic, political, recreational, public health—that’s what we have in this part of the world,” Dr. Havrelock said. About the presence of the Lakehead System, she added: “There’s no plan in which this looks good to anybody, except those who are running this out of convenience, and because they’re allowed to, because the public doesn’t know anything about it…It’s going on in plain sight, but invisible to most people because so many of the operations are concealed.”


Before 1859, there was no petroleum industry to speak of. There was crude oil in various places around the country—though nobody knew how much—and people used it in minor ways. Along Oil Creek near Titusville, in northwestern Pennsylvania, small amounts of crude would percolate. To collect it, people soaked blankets in pools of oily water, wrung them into a pan, and boiled the mixture down. Sometimes they could skim oil right off the surface of the water. In the 1840s, farmers along oil creek collected between two and twelve barrels a season, most of which was sold for medicinal purposes, to treat bruises, burns, and rheumatism.

Edwin Drake, a former train conductor whom some people called “Colonel,” knew it would be useful for other things. A chemistry professor at Yale named Benjamin Silliman indicated as much in an 1855 report, titled Report on the Rock Oil, or Petroleum, from Venango Co., Pennsylvania: With Special Reference to Its Use for Illumination and Other Purposes. Even before that, Samuel Keir was extracting kerosene from crude in Pittsburgh (kerosene could replace whale oil in lamps). Colonel Drake and his financiers, the Seneca Oil Company, thought there was a lot of money to be made in petroleum. They had heard crude oil would leak into salt wells, and thought maybe it would also leak more heavily into other holes drilled in the right places.

Historian Christopher F. Jones tells the story of oil’s beginning in Routes of Power: Energy and Modern America. In 1859, Drake tried for months to coax petroleum out of the ground. He used a steam engine to power his drill through bedrock, and supported the resulting holes with iron pipe to avoid a cave-in. He hired a saltwater driller named “Uncle” Billy Smith to run the machine. On August 28, a Sunday, Uncle Billy looked over the edge of a hole they had drilled and noticed oil. He lowered a can and scooped some out. Using a hand pump, Colonel Drake and Uncle Billy extracted gallons more and put it in a bathtub. They would need to get some barrels. Soon they would need about four hundred barrels every day.

Early on, pipeline companies had to post armed guards to protect their lines from sabotage.

Other people heard about Drake’s strike and dug their own wells. Many of them were successful. There were almost never enough barrels. It was common for people to strike oil and watch it run away across the ground because there wasn’t anywhere to put it. Some people dug wide holes around their wells to try to catch the crude, and were mostly unsuccessful. Oil soaked into soil, poured into rivers, caught fire and dismantled into the atmosphere. Even so, about eighty-five hundred barrels of oil were captured in 1859. Three years later that number topped three million.

Once they got the oil into barrels, producers would float it down Oil Creek to the Allegheny River and on to Pittsburgh, in the manner of lumber. Barges carried anywhere from twenty-five to a thousand barrels. Oil Creek was too shallow to support even the flat-bottomed boats, so people used pond freshets to move their product: they would put in a temporary dam to raise the water level, then pull it out and let the rush of water carry barges over shallower portions. Pond freshets might move a couple hundred barges at a time. There were a lot of wrecks. When people heard that a dam was going to be let go, they would crowd up on the riverbank to watch.

Transportation of oil in general was rough and unreliable. According to Jones, barrels were carted overland to Oil Creek, five or six at a time, by teamsters and their mules. Teamsters carried heavy “black snake” whips. Mules were run so hard they died after one or two years. Carts jostled fiercely on dirt roads and sometimes tumbled barrels off. About a third of the crude shipped down Oil Creek on barges leaked out or was dumped in a crash. In 1865, a lantern broke and started a fire that consumed eight thousand barrels of oil over one hundred barges. Some experts believe only fifteen or twenty percent of the oil extracted in the early Sixties ever reached the lamps of the country.

Soon enough, the railroads got involved. The Oil Creek Railroad, organized by oil producers, was first to connect Titusville to the grid. Eventually the major railroads showed up—the Pennsylvania, the Erie, and the New York Central. They carted oil in cars to refineries that had bubbled up in Cleveland, Pittsburgh, Philadelphia, Baltimore, and New York City. One Cleveland refinery was owned by John D. Rockefeller and his company, Standard Oil.

Standard Oil would go on to unbelievable success, a nearly ninety percent market share, and a high-profile antitrust case. One of the keys to Standard Oil’s dominance was cheap transportation. Railroads gave Rockefeller’s company rebates in exchange for its massive business, and for outside investments by Rockefeller that lowered railroads’ costs. Investigative journalist Ida Tarbell, in The History of the Standard Oil Company (1904), wrote that “It was the rebate which had made the Standard Oil trust, the rebate, amplified, systematized, glorified into a power never equaled before or since by any business of the country.”

People were already working on an alternative to rivers, mules, and railroads. By 1865, the first “gathering pipeline” had been built near Titusville, connecting a series of oil wells to a central point—in this case a railroad depot. Gathering pipelines were short, usually fewer than five miles long, and around two inches in diameter. The first successful long-distance pipeline was the Tidewater, which ran 110 miles from the Bradford Oil Fields to Williamsport, Pennsylvania, and was completed in 1879. It was constructed of wrought-iron pipes, six inches in diameter and sixteen to twenty feet long. Each section of pipe weighed three-hundred and forty pounds. It took eleven men to wrangle one into place. The lengths were fitted together using pipe tongs, which were a brand-new tool. No such pipeline had ever existed, so everything was custom designed.

Early on, pipeline companies had to post armed guards to protect their lines from sabotage. Railroads, worried about the appearance of the Tidewater, delayed and intentionally misplaced shipments of materials. The Northern Central and Buffalo, New York & Philadelphia Railroads pulled up pipes that ran under their tracks. Standard Oil published negative articles about the pipeline, purchased the refineries that were supposed to buy its oil, tried to buy a “north-south blocking line” of land to interrupt its path. The Tidewater went through anyway. Jones, the historian, wrote in Routes of Power that “oil entered the pipeline on May 28, moving at about the pace of a man walking.” Pumping stations drove the oil. Six days later it emerged in Williamsport. In the middle, it traversed the Allegheny mountains. When the oil arrived in Williamsport, people celebrated enthusiastically.

After Tidewater, pipelines were the rule. Standard Oil built a long one to Cleveland, and one to the Atlantic seaboard, at New York harbor. Other pipelines carried oil to Buffalo, Philadelphia, and Baltimore. One took off toward Chicago. Throughout the twentieth century, pipelines snaked around the Great Lakes and across the country. There are now more than a hundred and sixty thousand miles of oil pipeline traversing the United States, more than fifty-six thousand of them transporting crude. You wouldn’t know it from looking, because they run for the most part underground.

Since the 1950s, pipelines have usually been made from manufactured steel. Line 5, for example, is constructed of “seamless” pipe. Seamless pipe is extruded from steam billets to a desired length and joined with particular fittings on site. The walls of Line 5 and other underwater pipelines are coated on the outside with a layer of enamel, reinforced with fiber.

Some pipelines are still being built or expanded. In 2015, Enbridge completed an expansion of Line 67, the Alberta Clipper, which can now transport eight hundred thousand barrels of diluted tar sands oil per day. According to the Enbridge corporation, Line 5 does not carry, and has never carried, heavy crude oil of any kind. It carries eighty percent sweet or light crudes, some of them synthetic, and twenty percent natural gas liquids (NGLs). According to the Freshwater Lab, other pipelines and railroads in the Great Lakes basin are carrying diluted tar sands oil across rivers, lakes, and wetlands.

“There’s a kind of truism now that pipelines are the safest form of transporting oil, because a train could turn over, a truck could turn over, and they could literally catch fire and kill people,” Rachel Havrelock told me. “I certainly don’t want trains to catch fire and cause any deaths, but the whole thing is that we forget about the risks that pipelines do pose to our drinking water, and then we also forget about the other element, which is that their unseen nature means they can be neglected and mishandled.”


On June 25, 2010, Line 6B, a thirty-inch pipeline belonging to the Lakehead System, ruptured at a pumping station near Marshall, Michigan. Alarms sounded at the site. Diluted tar sands crude surged into Talmudge Creek, a tributary of the Kalamazoo River adjacent to the pumping station, and on into the Kalamazoo River itself. At a rate of one and a quarter miles per hour, it leached downstream in the direction of Lake Michigan, some 115 river miles away. Seventeen hours later, Enbridge responded by shutting off the flow of crude and notifying the U.S. Environmental Protection Agency. In a torrent of misfortune, rain deluged the streams, pushing oil into flood plains and thirty-five miles downstream. The EPA took over and restrained the spill before it reached any further. An estimated eight hundred and forty thousand gallons of diluted bitumen had escaped the pipeline. It was—and still is—the worst inland oil spill in U.S. history.

Between 2010 and 2014, more than 1.2 million gallons of oil were recovered from the waterways of western Michigan, and land nearby. The EPA has said there could still be around eighty thousand gallons mixed up in sediment.

Traditional methods of cleanup assume that the oil from a spill will float on water, and have developed accordingly. Booms are an example of this. There are two main kinds of oil booms: containment and absorbent. Both float near the surface of a body. Containment booms are usually plastic or metal, and create a barrier that corrals the slick as in a pen. Most of the containment booms I’ve seen are yellow, orange, or black. Absorbent booms are usually white at first, or sometimes blue, and are made of polypropylene wrapped in mesh. They are tubular in shape, and can link together like sausages. Absorbent booms soak up oil (or other hydrocarbons) and become brown. They are replaced periodically. For spills of more generous range, dispersants—chemical agents—can be deployed to scatter a spill. They form oil into droplets that can be pushed underwater, protecting surface species from grease and theoretically increasing the chance of oil breaking down naturally. Another technique is called “in situ burning,” or ISB, and proceeds as it sounds: crews build a barrier of inflammable booms and light the slick on fire.

Unlike most light or medium crudes, tar sands bitumen is more dense than water. After diluted bitumen spills into a river or other body, the diluting agent, which is usually extremely light, can vaporize, leaving the remaining bitumen to sink or suspend. At that point, booms, dispersants, and ISB are moot. When it gets to the bottom of a river or lake, bitumen mixes with sediment and has to be dredged. This is the best tactic anybody has found, though others have been tried. After the Line 6B spill, people stuck long poles into the sediment and stirred it up. They were hoping the crude would float to the surface. It didn’t, so they had to dredge the streams anyway. The whole process is described in detail in a long report released in 2016 by the National Academies of Sciences, Engineering, and Medicine. The report, Spills of Diluted Bitumen from Pipelines, was commissioned by Congress, and surveyed a cluster of research that sprung up in response to the Enbridge spill in Marshall, Michigan.

Between 2010 and 2014, more than 1.2 million gallons of oil were recovered from the waterways of western Michigan, and land nearby. The EPA has said there could still be around eighty thousand gallons mixed up in sediment. Enbridge paid $177 million in fines, and an estimated one billion dollars in cleanup costs. The U.S. government alleged that during the seventeen hours oil was pouring into Talmudge Creek, “Enbridge had restarted Line 6B on two separate occasions on July 26, 2010, pumping additional oil into the ruptured pipeline causing additional discharges of oil into the environment,” and that “the rupture and discharges were caused by stress corrosion cracking on the pipeline, control room misinterpretations and pervasive organization failures at Enbridge.” Later, Enbridge replaced the aging Line 6B with another, larger pipeline, also called Line 6B.

Many people are unaware of the Kalamazoo spill. It didn’t get as much national attention as you might expect. This is due partly to incredible timing in the news cycle: three months earlier, on April 20, the oil rig Deepwater Horizon had exploded, and the BP-operated Macondo wellhead had poured 3.19 million barrels of crude oil into the Gulf of Mexico.


On May 10 and 11, 2017, the Freshwater Lab held a summit in Chicago titled “Untrouble the Waters.” I went. The summit was free. Its title recalls the refrain from “Wade in the Water,” an old African-American spiritual, part of which goes like this:

Wade in the water
Wade in the water, children
Wade in the water
God’s a-gonna trouble the water

On the morning of the first day, Dr. Johari Jabir of UIC led an a-cappella group sing of a remixed “Wade in the Water.” Yuri Lane beatboxed. Every fourth line, instead of singing “God’s a-gonna trouble the water,” Dr. Jabir riffed on the themes of the conference—“we need to clean up the water,” etc. Maude Barlow, national chairperson of the Council of Canadians, a watchdog group, gave the opening keynote lecture. She said, “The Great Lakes are a public trust.” She said, “The waters of the Great Lakes belong to those who live on, around, and in them.”

The summit was organized into panels and working groups on water access, lead in city pipes, “watershed politics,” oil pipelines, human rights, environmental justice; one panel included five mayors and one former mayor from towns in the Great Lakes region. At the panel on pipelines, Chairman Robert Blanchard of the Bad River Band of Lake Superior Chippewa; two Michigan policy wonks, Liz Kirkwood and Jennifer McKay; Thomas Frank, an activist in East Chicago, Indiana; and Olga Bautista of Chicago’s Southeast Side Coalition to Ban Petcoke, talked about the complicated mixing of oil and water. I scribbled furiously on a yellow legal pad as they spoke, yanking sentences out of the air and pinning them to the page:

“What we’re seeing is a re-industrialization of the region.”

“Tar sands oil is the dirtiest of dirty fuels.”

“In our community, water is our life force.”

“Enbridge doesn’t have the capacity to deal with a major oil spill.”

“We’re talking seven hundred miles of coastline—the Coast Guard won’t be able to clean up a spill of that size.”

“The question is not if, but when pipelines will fail.”

“What we need is a comprehensive plan for decommissioning Line 5, and we need to organize around that.”

Enbridge, the U.S. Coast Guard, and other agencies have recently assured media they are prepared for managing a hypothetical spill from Line 5.

Everybody on the panel talked about Line 5. As Great Lakes pipelines go, Line 5 is the most notorious. Recently, activists and organizations have alleged that Line 5 is dangerously corroded, inadequately supported, operating over capacity, missing its protective coating, and generally in a state of neglect. In 2013, the National Wildlife Federation sent divers into the Straits of Mackinac to record images of Line 5. The video shows a pipeline coated in mussels and debris; long, unsupported sections of pipe; and broken supports. An October 2016 report, based on 2003 data, described “A number of areas where scouring effects from water currents caused sections of the pipelines to span freely above the bottom,” some of which were “in excess of the 75 foot limit” established in the initial easement. The longest unsupported section, on the easternmost of the two pipelines, reached 268 feet. Enbridge recently asked the state of Michigan for permission to install twenty-two new supports along the pipe.

Line 5 has never spilled oil into the Straits of Mackinac. But it has suffered twenty-nine spills elsewhere since 1953, totaling more than 1.1 million gallons. Enbridge has been responsible for more than eight hundred oil spills in the United States since 2000, including the 2010 Kalamazoo event and a second spill of two hundred and fifty thousand gallons that same year, just southwest of Chicago. And with a record like that—well, people talk.

I was curious about a number I’d heard tossed around at the pipeline panel: “700 miles of coastline.” It comes from a 2016 report out of the University of Michigan, which found that “more than 700 miles of shoreline in Lakes Michigan and Huron and on their islands are potentially vulnerable to an oil spill in the Straits.” Researchers ran simulations for oil spills at different locations and times of year, using a hydrodynamic model. “More than 700 miles of shoreline” includes a variety of hypothetical spills, and is a worst-case number. A single case could affect a hundred and fifty miles or so. The report also suggested that fifteen percent of Lake Michigan and sixty percent of Lake Huron were at risk in a Line 5 rupture.

Enbridge, the U.S. Coast Guard, and other agencies have recently assured media they are prepared for managing a hypothetical spill from Line 5. Enbridge, for its part, has further said that it monitors the pipeline vigorously, with pressure gauges and fly-overs and other advanced techniques. The company is also developing an autonomous vehicle, in partnership with Michigan Technological University, which would be able to inspect the four and a half miles of Line 5 that live underwater. But many people are not convinced that the ecological future of a region ought to be entrusted to an energy delivery company and its sixty-four-year-old pipeline. One of them is Dr. Rachel Havrelock. Five others were on the Freshwater Lab’s pipeline panel in May.

Some of them, evidently, also rank highly in the government of Michigan. In 2015, the state published a report in response to events including the 2010 Kalamazoo spill. The Petroleum Pipeline Task Force Report, presented by Michigan’s Department of Natural Resources, Department of Environmental Quality, Department of Transportation, and Attorney General, among others, addressed pipelines in general and Line 5 in specific. It recommended that no “heavy crude” oil be transported through the Straits of Mackinac, and that an independent analysis of alternatives to Line 5 be carried out.

Two years later, in June 2017, a draft of the independent analysis was released for public comment. The most promising alternatives, from the perspective of the report, included the construction of a new pipeline, either slid through a trench or buried entirely underground. Costs of such a replacement were estimated in the area of $180 million. Railroads and the use of existing infrastructure were ruled out entirely. A few weeks later, five members of the Michigan Pipeline Safety Advisory Board published a dissent, claiming there was missing data, and that the document failed to, among other things, “accurately assess spill risk” and “properly evaluate impacts of a spill.” Decommissioning of Line 5—the preferred route for many activists—was included in the analysis at a cost of $200 million.


If you hang around pipeline conversations, one word you will hear over and over is “risk.” The classic definition of risk is the potential for an unwanted outcome. To calculate risk, scientists use the formula Risk = Probability x Loss (or consequence). Usually in the case of oil pipelines this is couched in terms of ‘Probability of Failure’ and ‘Potential Damage.’ For the most part, the pipeline industry emphasizes the former: oil pipelines are less likely to spill than rail cars and semi trucks. And some stretches of pipe—like the four and a half miles of Line 5 under the Straits of Mackinac—have never ruptured. Pipeline objectors focus on the other side of the equation: what would happen if they did. It’s only a matter of time, they say, until the next Line 6B.

“This system is the most extraordinary freshwater system that exists on earth…There is nothing more precious than fresh water, by anyone’s measuring stick.”

Of course, there will never be another incident exactly like the one on Line 6B in 2010. Catastrophe never repeats itself. The conditions of the world are such that each occurrence unfolds in singular fashion, directed by the precise circumstances of its appearance. What will happen—and where—is anyone’s guess, and perhaps none of us are right. Some people hazard qualified predictions using the best information and technology available, but at any moment, a heavy rain could come along and sweep all these projections downstream. This is one reason researchers often publish a range of possible outcomes—there is always another variable to be considered. One of the greatest threats to the stability of Line 5, according to recent reports, is—of all things—a stray anchor, dragged across the bottom of the Straits by an unwitting vessel.

“This system is the most extraordinary freshwater system that exists on earth,” Rachel Havrelock said of the Great Lakes when I spoke with her in May. “Whether you care about human health and longevity, or whether you care about resources of value and their economic potential, or whether you care about the viability of ecosystems, it all matters here. There is nothing more precious than fresh water, by anyone’s measuring stick.”

A lot of what we have to think about, as people, can’t be adequately expressed with numbers. Oil spills are like that. How to quantify the cost of a rupture in the Great Lakes? Or, for that matter, outside of them? At what number of barrels—what precise number—do we cross from “OK” to “disaster”? Yet the computation of risk requires that we furnish an answer. As long as fossil fuels remain the power source for civilization, pipelines will be pumping them around the earth and its water. And we will be left assessing the consequences of a spill—calculating the incalculable.


*Correction: an earlier version of this story misstated the year in which Line 5 was constructed as 1853. The pipeline was constructed in 1953.



Ryan Schnurr is a writer from northeast Indiana. His first book, In the Watershed, is forthcoming from Belt Publishing in October 2017.

Cover image courtesy the UIC Freshwater Lab.