In the winter of 1969, a drill runner stood at the face of a tunnel beneath the Continental Divide in Colorado. He was not above the mountain, but inside it, roughly two thousand feet deep. The granite and gneiss around him had formed about 1. 7 billion years ago, under conditions that had nothing to do with the project’s schedule but everything to do with its difficulty.
The rock was as hard as billion-year-old rock gets—harder than the steel drill bits that chewed into it, harder than the air hammers driving those bits, harder than any force the project’s equipment could apply without wearing itself out in the attempt. The face stood about fourteen feet high and forty-two feet wide. It was covered with drill holes, a hundred or more, laid out in a pattern designed by the head driller to direct the energy of the explosive charge outward toward the tunnel perimeter rather than inward toward the depth and the men who would enter that space in the hours after the blast. The mountain above him consisted of roughly fifteen hundred feet of rock and soil at the deepest point over the tunnel’s centerline.
The portal he had entered that morning was behind him, invisible from where he stood. The air at eleven thousand feet above sea level contains about thirty-four percent less oxygen than the air at sea level. A man breathing that air while drilling rock under pressure and at a fast work pace creates a deficit between what his body needs and what the air provides. The air inside the tunnel was colder than the air outside, and the air outside was already freezing in December in the Colorado Rockies, because the mountain was absorbing heat from everything that entered it.
His breath was visible. The steel of the drill rod was cold enough to hurt bare hands. He had been underground for hours. He would remain underground for hours more.
This is the story of how Colorado built a tunnel through the Continental Divide, the backbone of North America, at an elevation where the air was thin enough to confuse experienced workers and cold enough in winter to freeze diesel fuel in construction equipment before the day shift began. It is the story of a project engineers had studied since the 1930s and been told was too expensive and too geologically uncertain, until the economics of the Interstate Highway System, the booming ski industry, and a western slope that could no longer wait for Loveland Pass to reopen made not building it a decision no administration could defend. It is the story of the men who drilled, blasted, planned, and supported 8,941 feet of tunnel through a mountain of Precambrian rock from both ends at once; who breathed air a third less dense than what their lungs were designed for; who worked alternating shifts inside the rock so that the excavation never stopped; and whose names do not appear on the memorial plaque at the tunnel’s east portal. The tunnel is named for Dwight David Eisenhower.
It was built by men whose names appear in the payroll records of the contractors who hired them. To understand why Colorado spent three decades debating whether to drill a tunnel straight through the Continental Divide, you have to understand what the Divide is and what crossing it cost. The Divide runs in Colorado roughly north to south down the middle of the state along the high ridgeline of the Rocky Mountains. Water falling on the eastern side flows toward the Atlantic basin via the South Platte, Arkansas, and Rio Grande rivers.
Water falling on the western side flows toward the Pacific via the Colorado River and its tributaries. The Divide is not a single obvious mountain peak. It is a succession of high passes and high terrain, and crossing it by land means finding a route through or over ground rising anywhere from about eleven thousand to nearly fourteen thousand feet, depending on the route you choose. Denver sits on the eastern slope of the Rockies at roughly 5,280 feet.
The western slopes of Colorado—the Grand Valley, the Roaring Fork, the Gunnison country, the plateau lands and beyond—lay on the other side of the Divide, and in the 1950s and early 1960s they were economically isolated from Denver in a way determined by the state’s geography and that no road improvement short of a tunnel could fundamentally change. The road connecting Denver to the western slope ran over Loveland Pass. US Highway 6 over Loveland Pass reached 11,990 feet at the summit. It was two lanes, steep and winding, and vulnerable to closure in winter storms that could arrive without warning and shut the pass for hours or days at a time.
A truck carrying goods from Denver to Grand Junction in January of 1963 went over Loveland Pass because there was no other way. The pass closed on average a few dozen days every winter in the years before the tunnel opened. Every closure meant cargo waiting. Every closure meant a driver stuck on the wrong side sat in whatever shelter the pass provided, which was not much.
Chain laws were enforced whenever conditions permitted travel at all. A loaded semi descending Loveland Pass on icy conditions with chains on its drive wheels was handling a weight, a grade, and a set of curves no road engineer had designed for the equipment that would eventually use it. Accidents on Loveland Pass in winter were not incidental events. They were an expected feature of Colorado’s winter shipping operations.
Beyond freight, the western slope was where the ski industry had begun generating economic activity on a scale the prewar highway surveys had never anticipated. Vail Mountain opened in 1962. Arapahoe Basin, Breckenridge, and Keystone were already in operation. The traffic demand these areas generated from Denver on winter weekends was demand a two-lane road over an 11,990-foot pass could not handle, and in bad weather could not handle at all.
Skiers who drove from Denver to Vail in 1964 crossed Loveland Pass because there was no other way, and did so in the same conditions that made the pass dangerous for truckers, except without chains. The people who calculated the cost of the pass to the state were the engineers, transportation planners, and highway department officials whose job it was to model traffic and assess the cost of delay. Their numbers, through the late 1950s and into the early 1960s, produced a consistent conclusion. The demand for a direct crossing of the Continental Divide near Loveland Pass was real and growing.
The question was not whether the tunnel was justified. The question was whether the geology of the mountain would allow it and whether the financing could be arranged. The first serious engineering study of a tunnel through the Continental Divide near Loveland Pass was conducted in the late 1930s. The engineers who studied it reported what engineers always report when studying large tunnels through mountain ranges.
The geology was variable, the rock hard in some sections and broken in others. Water inflow was likely. The cost would be high relative to the traffic volumes of that era. The project was deferred.
Through the 1940s, the war consumed the capital and engineering attention a tunnel project required. Through the early 1950s, economic conditions were not favorable enough to move the proposal forward. The pattern repeated itself. Engineers studied the situation.
Engineers reported that the mountain was difficult and the cost high. Politicians read the report and declined to act. What changed in the late 1950s was the Federal-Aid Highway Act of 1956, signed into law by Dwight Eisenhower, which created the Interstate Highway System with federal funding at a ratio that changed the cost structure for projects the states could not finance independently. Interstate 70 was in the original network plan, crossing Colorado.
In the original plan, I-70 crossed the Continental Divide at a location that required a tunnel or a route adding significant distance. The federal participation changed the equation. A project beyond Colorado’s capacity to finance on its own became a project the state could fund, with the federal share covering the majority of construction costs. The Colorado Department of Highways engaged engineering firms to study the tunnel alignment in detail.
The engineers drilled exploratory boreholes into the mountain at intervals along the proposed centerline, sent samples to laboratories to determine the mechanical properties of the rock at various depths, and mapped the geology of the Divide near the proposed portal sites. They identified fault zones and zones of fractured rock. They identified areas where groundwater moved through the mountain and calculated the volume of water the tunnel would intercept during construction. They analyzed methane content in core samples, because this mountain contained zones where organic material had been incorporated into Precambrian rock by the geological processes that formed it.
The presence of organic material in the rock meant the possibility of combustible gas in the excavation during construction. The engineers’ conclusion was that the tunnel was feasible. The geology was not uniform and not ideal, but it was not disqualifying. The Precambrian granite and gneiss forming most of the mountain was hard and would require significant blasting energy to advance the excavation, but hard rock requiring drilling and blasting was manageable in ways that soft, fractured rock was not.
The fault zones and fractured sections the borings identified would require special ground support. The water flows would require continuous pumping. The gas would require ventilation and continuous monitoring. None of these problems was new.
They were standard problems in deep mountain tunneling and had been solved before in other mountains by other crews. The length, 8,941 feet, was not the longest tunnel in the country. It was longer than most highway tunnels in the United States at the time. The elevation, 11,013 feet at the east portal, was unprecedented for any vehicular tunnel anywhere in the world.
No highway tunnel on earth had ever been excavated at that altitude. And the altitude was not merely a record to be entered in a reference book. It was an engineering condition that affected every aspect of the work. Air at 11,013 feet contains roughly sixty-six percent of the oxygen in air at sea level.
A diesel engine operating at that density loses a proportionate share of its rated power because it cannot burn fuel at full efficiency without sufficient oxygen. An air compressor powered by diesel at 11,000 feet produces less compressed air than the same compressor at sea level. A drill running on less compressed air drills more slowly. A man coming from Denver needs two to three weeks to fully acclimatize before his aerobic capacity returns to near sea-level values.
A man doing hard work in his first days at altitude performs at a fraction of his usual capacity without always realizing it, because the altitude impairs the judgment that would have revealed the impairment. Funding was arranged through a combination of federal highway money and state highway appropriations. Contracts were let. Work began in 1968.
The workforce that arrived at the Eisenhower Tunnel portal camps in 1968 was a specialized one. Tunneling in hard rock was not a job a general laborer could move into from another classification without training. It required the skills of drilling, blasting, and mucking in a confined space; of reading the rock face to anticipate what the next round would produce; of designing the drill pattern to direct blast energy toward the muckpile rather than the roof; of knowing the right quantity of explosive for the rock type. These were skills learned in mines and previous tunnel projects, carried by workers who moved from one tunnel contract to the next across the western United States.
They came from Colorado’s established mining communities—from Leadville, from Creede, from the San Juans. They came from water and highway tunnel projects in California, Nevada, and Utah that preceded this one. They came from wherever the last tunnel had ended, when that tunnel’s contract ran out and the workforce scattered to the next job. They were paid wages that reflected the premium that hard-rock tunnel work commanded over general construction labor, because the risks were higher, the skill requirements greater, and the conditions worse than surface construction by every measure.
They were unionized. They understood the value of their labor in the existing market and bargained accordingly. The premium classifications—driller, powderman, drill runner’s helper—were held by men whose seniority in the trade entitled them to those jobs. The workforce on the Eisenhower Tunnel project was predominantly white in the premium classifications, reflecting the composition of the hard-rock unions of the American West and the history of those unions and the industries they came from.
The men in the labor classifications outside the skilled trades were more diverse. The payroll records contain the details. The project operated on three shifts, advancing the work around the clock, because the construction schedule and the public demand for a crossing that did not close in January did not permit a workforce that stopped at sundown. A heading that advanced for sixteen hours and rested for eight fell behind a heading that advanced for twenty-four.
A delayed tunnel excavation cost the contractor money, wasted the state’s time, and cost drivers on Loveland Pass another winter of suffering on the pass. The work in the tunnel ran around the clock. Working conditions inside the tunnel were unique, with no equivalent in surface construction. The temperature inside the mountain held at about forty degrees Fahrenheit regardless of season, because the mountain maintained a thermal equilibrium that the seasons above could not penetrate.
Men entering in July and men entering in December faced the same temperature. The cold was not variable. It was the permanent condition of the rock. The men working at the face dressed in layers suited to forty degrees while expending high physical effort.
During periods of heavy exertion, those layers were not enough. During periods of standing by while the air cleared after a blast, or waiting for the survey crew to finish their readings, the layers were barely enough. The air inside the tunnel was whatever the ventilation system provided. In the early months, when the heading was close to the portal, natural ventilation through the tunnel could provide adequate air exchange.
As the heading advanced and the distance from the portal increased, natural ventilation became inadequate. The blast that advanced the excavation produced nitrogen dioxide, carbon monoxide, and particulate matter in concentrations that did not allow men to return to the face until the air was changed. The ventilation system, consisting of ducting running from the portal to the heading and fed by fans mounted at the portals, was the system that made re-entry possible after a blast. The fans ran continuously.
If the fans stopped, the work stopped. The dust from the drilling was silica dust, airborne from the fractured granite and gneiss. Silica dust at concentrations above the permissible limit causes silicosis, a progressive and irreversible fibrosis of lung tissue that reduces respiratory capacity over years of exposure, with no cure other than attempting to slow its progression. Water sprayed on the drill bits and on the muckpile during cleanup was the primary control measure.
Wet drilling generated less airborne dust than dry drilling. So wet drilling was the adopted practice. The water used to suppress dust seeped to the floor of the tunnel and pooled there, and had to be pumped back to the portal for discharge. The men worked in a space that was wet underfoot, cold, dusty, and filled with the sound of air drills hammering against granite at full speed.
Sound levels inside a tunnel excavation in a confined space are levels that damage hearing even at exposures shorter than a full shift. The tunnel jumbo—a multi-arm platform that put multiple air drills against the face at once—could run ten drills or more simultaneously. Ten air drills working Precambrian granite in a space forty-two feet wide and fourteen feet high created a noise environment that hearing protection could only attenuate partially, not eliminate. The men who dug the Eisenhower Tunnel for months wore whatever hearing protection the industrial safety practices of that era provided.
The practices of that era did not meet current standards. The methane gas the geological surveys had identified was encountered during construction. The mountain contained flammable gas in fractured zones along fault lines and in sections where organic material had been incorporated into the rock. When the advancing excavation intersected a gas-bearing zone, gas seeped into the tunnel.
Monitoring equipment detected the concentration. Ventilation protocols were designed to dilute it below the explosive threshold. These protocols worked as designed, until they did not. In October 1970, a penetration of a gas pocket during excavation released gas into the tunnel at a concentration the ventilation system could not dilute quickly enough.
An ignition source, attributed by investigations to a spark from equipment, ignited the gas. The explosion and resulting fire killed two workers and injured others. It halted the project for weeks. While investigations proceeded, the ventilation system was redesigned, and gas monitoring protocols were revised and re-implemented.
Excavation resumed. Gas was expected to be encountered again. It is what always happens in a mountain containing such gases. The altitude affected everything the men did and everything the equipment did at every stage of the project.
A driver coming from Denver to the tunnel portal gains vertical elevation in a single trip that physiologists measure in acclimatization time. The first days at altitude caused headaches, fatigue, and reduced aerobic capacity. A man moving muck or operating a drill at altitude before his body acclimatized to the lack of oxygen worked at a fraction of his usual sea-level capacity without always realizing it. Project management addressed acclimatization by housing workers at the tunnel portal elevation, so they did not have to commute across the elevation differential between shifts.
Whether altitude sickness monitoring was comprehensive, and whether workers showing symptoms were pulled from the heading in a timely manner, is a matter in the project’s medical records rather than in any widely available public report. The method used to build the Eisenhower Tunnel was drill and blast, the method used for hard rock tunneling since the development of reliable high explosives in the nineteenth century, and the method used in this mountain because the rock was too hard for any other method to be economically viable. The jumbo advanced to the face. The drillers set the pattern and drilled the round—a set of holes with geometry designed by the head driller based on the rock conditions and the blast design.
The cut holes were drilled in the center of the face at angles to create the initial cavity. The easer holes widened that cavity outward. The perimeter holes defined the excavation boundary and were loaded with reduced explosive charges to limit damage to the rock the tunnel would rely on. When drilling was complete, the crew withdrew from the working face.
The powderman crew loaded the charges in the sequence the design specified. Detonator circuits were extended back to a firing station. The tunnel was evacuated and checked for personnel. The blast was fired.
The blast fractured and displaced a section of rock several feet deep across the entire face. The blast gases and particulate matter required the ventilation system to run at full capacity before re-entry was permitted. When the air cleared and gas monitors confirmed safe levels, the mucking crew entered with their loaders and moved the broken rock to the tunnel floor where it was loaded onto rail cars or trucks and hauled to the portal and dumped on the waste pile outside. After mucking, the ground support crew entered to install rock bolts, wire mesh, and shotcrete—the immediate support that prevented the freshly broken tunnel roof from collapsing before the permanent concrete lining was placed.
Once support was installed, the jumbo advanced to the new face. The round had advanced the tunnel eight to twelve feet depending on rock conditions. The cycle repeated on the next shift. Three shifts running this cycle continuously could advance the tunnel fifteen to thirty feet per day in hard rock.
In fault zones, where rock was broken and water flowed and gas readings required extra ventilation time before re-entry, progress slowed. Support requirements increased. The time consumed by bad rock did not show up as weather delays in the contractor’s records. It showed up as lower advance rates, higher cost per foot, and pressure on the schedule the project demanded.
The tunnel was excavated from both ends simultaneously, the east portal and the west portal, with the two headings advancing toward each other through the mountain. Excavating from both ends cut the total time required by roughly half compared to starting from a single heading, but it imposed a surveying problem: ensuring that two tunnels advancing independently, separated by thousands of feet of rock with no line of sight between them, met within the specified alignment tolerance at the midpoint. Survey control was managed through an optical control network extended from surface monuments of known position down through the portals and along the tunnel by licensed surveyors who checked and rechecked the alignment at every advance. Any deviation in excavation of fractions of a foot per hundred feet of advance accumulated over thousands of feet into a misalignment large enough to miss the oncoming bore.
The survey crews measured. The drill crews drilled to the coordinates the surveyors provided. The surveys were precise. The two headings met in 1972.
The breakthrough—the moment when the remaining rock between the two headings was drilled, blasted, and mucked away so that it became possible to see through the tunnel—was the moment the project had been working toward since 1968. The survey tolerance at the point of breakthrough was within the required specification. The tunnel was fully connected. The concrete lining, roadway surface, ventilation equipment, and traffic control systems were completed through the winter of 1972 and into early 1973.
The Eisenhower Tunnel opened to traffic on March 8, 1973. The opening ceremony included Governor John Vanderhoof and state and federal highway officials. Photographs from the ceremony show officials at the east portal. The first vehicles to traverse the tunnel were driven by officials and dignitaries who had not done the excavation work.
The tunnel is named for Dwight D. Eisenhower, who signed the Federal-Aid Highway Act of 1956 and died in 1969. He never saw the tunnel bearing his name. The second bore, built through the mid-to-late 1970s and opened in 1979, is named for Edwin C.
Johnson, a Colorado senator who had championed the tunnel project in the 1950s and did not live to see its opening either. The men who excavated the tunnel are not in the photographs from March 8, 1973. The workers who mucked the muck, set the rock bolts, ran the ventilation ducts, monitored the gases, drove the supply equipment, and maintained the jumbos are not named on the memorial plaque at the east portal. The two workers who died in the explosion and fire of October 1970 are recorded in the project’s injury and fatality documents and in the records of the Colorado labor and safety agency that collected such reports in 1970.
Their names are not among the public commemorations of the tunnel. The tunnel is named for a president and a senator. The men who bored it through the mountain exist in the payroll records of the contractors who paid them. The Eisenhower–Johnson Memorial Tunnels carry I-70 in both directions under the Continental Divide.
The posted speed limit inside the tunnel is fifty miles per hour. Crossing the full length of the tunnel at that speed takes about two minutes. The ventilation system runs continuously because 8,941 feet of enclosed interstate highway carrying traffic produces carbon monoxide in concentrations requiring active air exchange in all operating hours. The fans at the portals run every day the tunnels are open, which is every day, because Loveland Pass, which the tunnels replaced, still closes in winter while the tunnels do not.
The inspection program sends crews into the tunnel on scheduled cycles to inspect the concrete lining, assess the condition of the rock support systems behind the lining, and evaluate the drainage systems handling the groundwater that still enters the mountain. The mountain has not stopped pushing water into the tunnel since the first heading intersected the first water-bearing zone in 1968. The drainage systems handling that water have run continuously since then. The rock bolts at the crown of the tunnel still carry the load of the rock above them.
The concrete placed in the winter of 1972 is the concrete vehicles pass beneath today. At the center of the tunnel, under the highest point of the arch, the Continental Divide passes through the mountain roughly fifteen hundred feet above. Water falling on the surface at this moment divides. Rain on the eastern side of the line above this point drains toward the South Platte River.
Rain on the western side drains toward the Eagle River and then the Colorado. That division happens in the rock above the concrete placed by the crews of 1968 to 1973 to keep the mountain out of the road. The rock they excavated lies in the waste piles at the portals, where it has settled since the cleanup crews loaded and hauled it out. The tunnel they advanced at rates up to twelve feet per day through rock formed before complex life existed on this planet is the tunnel you drive through now.
The men who excavated it breathed air a third lighter than the air at the base of the mountain. They drilled at forty degrees Fahrenheit in the permanent cold of Precambrian rock. They loaded charges into granite that had been in place for 1. 7 billion years, and blasted it, and broke it, and loaded it into trucks.
Two of them died when the mountain took back what it was holding. The rest finished the contract and moved on to the next one. When the heading was advancing and the survey calculations indicated the oncoming bore was drawing near, the head driller could sometimes hear the sound of the drills on the other side of the rock. He could not see them.
He could hear them through the face. The sound of air hammers on steel carried through a billion years of granite from the tunnel advancing toward him. He drilled where the surveyor told him to drill. He fired when the supervisor told him to fire.
The rock came out. The other heading appeared. The Continental Divide did not move.
The tunnel is still there.
