Fire on the Horizon

by Tom Shroder

  That familiar, nagging twitch of concern had no relation to what gripped her now. She came close to telling Dave what she was feeling and begging him not to get on the plane, but that would have been crazy, she knew. She didn’t believe in this. She refused to believe in it.

  The sick feeling persisted. It stayed with her on the lonely drive home, swirling around her in a black fog until Dave called the next day, April 15, to say the helicopter had landed on the rig. She could feel her muscles relax. Now he was safe.

  CHAPTER FOURTEEN

  POSITIVE TEST

  April 19, 2010

  Block 252, Mississippi Canyon, Gulf of Mexico

  Jimmy Harrell didn’t like what he was hearing.

  He was at the 11 a.m. meeting on April 19. They had been running the final long string of casing down into the hole since the day before, and now they were looking ahead to the final significant task before leaving Macondo behind—the cement job that would seal off the bottom of the well. Ronald Sepulvado, the senior BP company man he had worked with for years, had been called to shore to participate in a mandatory well control school to refresh his knowledge of how to deal with a well that threatened to go out of control. Sepulvado’s temporary replacement was Bob Kaluza, a man who’d been flown to the Horizon from BP’s biggest asset, the Thunder Horse PDQ production rig.

  Kaluza had plenty of experience, thirty-five years in the oil field, but having spent just a few days aboard the Horizon, Sepulvado’s notes could not have adequately conveyed Transocean culture, the personalities of the crew, the specifics of the equipment the Horizon used, or how to handle operations on an asset not owned by BP. But here he was nonetheless, outlining to the Horizon’s drill crew the plan for the upcoming cement job.

  Something he said took Jimmy by surprise, and taking Jimmy by surprise on a rig was not an easy thing to do. Rigs had been his life for more than three decades, and he’d worked his way though just about every job a man could have on the rig, beginning with deckhand, and he’d seen just about everything there is to see.

  Now he was the OIM, the boss, and the senior drilling hand aboard. Like so many in the Gulf, he was a southerner, born in Mississippi, and a true gentleman, careful to call the men in positions above him, even men years his junior, “Sir,” and all women, regardless of age or position, “Ma’am.” His round face, soulful eyes, and droopy white mustache made him seem approachable, like a favorite uncle. Everyone just called Jimmy “Jimmy,” or “Mr. Jimmy” if they were a junior hand and felt presumptuous addressing the boss by his naked Christian name. But Jimmy wouldn’t have minded if they had. He had the kind of self-confidence that expressed itself as extreme calm in difficult situations. He had been in enough tight squeezes in his life to know that if you kept an even keel, tough times usually ended up for the best.

  He had needed that faith on a well like Macondo, which had been a true bitch of a project right up to the end. Now this cement job would all but finish it up.

  What caught his attention now, and kind of shocked him, was the company man saying they intended to run special, and more expensive, nitrogen foam cement for this job.

  Let me stop you right there, Jimmy said. He knew a thing or two about nitrified cement. He was no stranger to the stuff; they used it all the time—they’d used it earlier on Macondo. But they’d always used it up near the top of the well, at the muddy sea bottom. In his more than three decades in the business, he’d only seen it used in a deep part of the well twice.

  The problem was that at great depths—and this cement was going down three and a half miles—the high pressure threatened to squeeze the nitrogen bubbles right out of the mix, which would then rise up the well, creating tiny holes and channels that would render the cement job worthless as a seal against oil and gas. Jimmy didn’t like this at all, and he didn’t even know about the tests Halliburton had been running on the proposed cement slurry—so far, the mix had failed to prove itself stable.

  His objections came from the gut, and his long experience. But Kaluza wasn’t giving in. Lighter cement, less likely to break through the well’s fragile walls, was essential to the plan. The nitrogen had to stay.

  Jimmy figured he wasn’t going to win this argument. The company man was the customer, the man who wrote the checks. This was BP’s project. But it was Jimmy’s rig, and he wasn’t going to let it drop without at least a little razor-edged sarcasm.

  “Well,” Jimmy said, “I guess that’s what we got those pinchers for.”

  The cement job began at 8 p.m. on April 19. The first step was to make sure the hole was clean and safe and the casing ready to adhere to the cement. For this particularly delicate job, the standard practice called for a full bottoms-up circulation, achieved by pumping enough clean well fluid down the hole to push all the existing mud out the top.

  This can be critical to getting a good cement job. Drilling mud’s tendency to gel when sitting idle is a useful property. If it remained entirely liquid, whenever the pumps stopped during drilling, all the cuttings and debris the mud had been pushing out of the hole would just sink back down to the bottom. The gelled state is thick enough to hold the debris in suspension until pumping resumes. But that very property can introduce serious problems. Vigorous circulation of clean mud will break down the clots and push them out of the well. If the circulation is too gentle or too short, some of the clots will remain clinging to walls, especially where the casing is squeezed against the wall of the well—exactly the scenario Jesse Gagliano had warned about.

  Incoming cement seeks the path of least resistance. It will flow around the clots rather than displace them, or force the clots to finally dissolve, only to contaminate the cement, riddling it with channels that can be invaded by pressurized oil and gas.

  But BP elected not to do a bottoms-up. Doing a full circulation may have threatened to fracture Macondo’s fragile walls. The pressure of the pumping had been reduced to avoid that, but reduced pump pressure also meant that doing bottoms-up would take hours of rig time—possibly more than BP was willing to budget.

  After circulating only 342 barrels of mud, about half the amount that would be needed to do a bottoms-up, Halliburton contractors began to pump the cement.

  They actually began by pumping a lighter-than-water base oil. This was yet another way to address the fear that the cement job would push out the well’s bottom again. Even the lighter foamed cement was more than twice as heavy as the base oil. Stacked up in the annulus for 1,000 feet, the cement, plus 12,000 feet of drilling mud on top of it, would have added up to tons of pressure bearing down. Pumping in the base oil would displace an equal volume of drilling mud out the top of the riser, significantly lightening the overall load on the well’s bottom.

  After the base oil went in, followed by spacer fluid to prevent contamination from the oil, they began to pump in the cement itself. The amount had been carefully calculated and measured so it would fill 190 feet in the bottom of the final casing, then get pushed up, as gently as they dared, from the bottom of the hole into the annulus, reaching a thousand feet toward the surface. When all the cement was in place, they checked the returns in the mud pits. The drill crew carefully calculated the amount of mud displaced. It was a match for the amount of cement that went in—what they called “full returns.” This meant that all the measures they had taken had succeeded in preventing the well walls from giving way.

  But it didn’t say much, if anything, about whether cement had formed channels, as predicted by the model. The cement could have channeled like crazy or failed to adhere to the casing and still give full returns.

  The only way to determine whether the cement was solidly in place would be to let the Schlumberger crew that had been waiting on board do the cement bond log test. They would run their delicate instruments down into the well, dangling from a hook at the end of a wire cable, unspooling from a huge reel and making thermal and sonic measurements as they went. Computer analysis of the results would create a 360-degree imag
e of the cement in the annulus, revealing any voids the cement had failed to fill.

  But the BP engineers, both on the rig and in Houston, saw the full returns and declared the cement job a success. The Schlumberger crew would be sent home on a morning BP flight off the rig without conducting the bond log test, saving BP $118,000 and the half day it would have taken to run the test—amounting to roughly another half million dollars saved.

  A few days earlier, the nitrogen foamed cement had been tested in a lab to determine how long it would take to set under the anticipated temperature and pressure conditions that would exist at the well bottom. The lab test revealed that after twenty-four hours, the foamed cement still hadn’t set. It took forty-eight hours to reach maximum hardness.

  But just short of eleven hours after pumping liquid cement into the bottom of the well, BP officials ordered what is called a “positive pressure test” to establish the integrity of their new long string casing. They closed off the well at the BOP in preparation, then pumped mud down the “kill line,” which skirts around the annular preventer and can increase pressure at the bottom of the well even when the annular is closed. They kept pumping until 2,500 pounds per square inch of pressure built up inside the newly cemented casing. They held the pressure at that level for thirty minutes. If the pressure eased in that time, it would mean there were leaks.

  The pressure held steady for the full half hour, and the test was deemed a success.

  The fact that the positive test was conducted when the cement may not have yet been fully hardened could conceivably have caused a new problem. Pressuring up the casing makes the steel expand slightly. If the cement mix was still semifluid, the expanding casing could have broken the bond between steel and cement, leaving a small but potentially lethal space that gas could pass through on the way to the surface.

  The cement job and the testing had gone on through the night, into the morning.

  It was Tuesday, April 20, 2010.

  CHAPTER FIFTEEN

  NEGATIVE TEST

  0630 Hours, April 20, 2010

  Block 252, Mississippi Canyon, Gulf of Mexico

  Miles “Randy” Ezell, the Horizon’s senior toolpusher for four years and one of the original crew who brought the rig out of Korea, had his first meeting just as the sun broke the wavering curve of the Gulf. It was 6:30 and uncommonly calm with the promise of a beautiful subtropical spring day. The first meeting was a tele-conference review of the coming day for Transocean management. That meeting rolled into a meeting he attended with the BP company men, conferencing in with their people in Houston to discuss the plan for the day, followed by the Horizon supervisors’ meeting with all the department heads.

  The department head meeting got going around 8:30, and it lasted for some time, mostly as they discussed the DROPS program, which was designed to curtail injuries due to one of the bugaboos of rig life—dropped objects. Nobody needed to tell them about the dangers of gravity in such a compact and intense industrial zone as the Horizon. Anything dropped off the 240-foot-tall derrick—a bolt, a hammer, even a pencil—could became a deadly missile. And a load of steel casing dropped overboard during transfer from a workboat might just take out the riser or even damage the BOP, a multimillion-dollar mistake. Transocean attacked the problem in typical fashion—with an acronym, a set of principles, and an aggressive program of proselytizing to drive those principles home. This included signs, literature, tons of required documentation and forms to fill out, and, of course, meetings. The Horizon had a record and a reputation to protect. The rig had kept serious accidents to an admirable minimum. This very day, four VIPs from Transocean and BP were flying in to deliver an award to the Horizon for controlling the number of “lost time” incidents, defined as any accident resulting in a worker missing time from work beyond the day or shift the accident occurred. In fact, there had been no lost-time incidents on the rig for seven years. And as everyone in the offshore industry knew so well, lost time was lost money.

  When the DROPS meeting was finally over, Randy had just enough time to make rounds of the rig floor before yet another meeting, this one the regular “pre-tour safety meeting,” essentially a huddle for the oncoming crew to discuss plans for the day, led by Jimmy Harrell, the driller Dewey Revette, Randy himself, and some others. He walked into the meeting room near the galley as Dewey outlined the order of battle.

  Today was the day they would begin closing up the well for “temporary abandonment,” until a production platform was designed and built and installed above it. They had just put in place a cement plug at the bottom of the well. Now all they had to do was set another cement plug near the top. Federal regulations require that “top plugs” be installed no more than a thousand feet below the seafloor. But BP had received approval to put Macondo’s top plug deeper than that, 3,300 feet below the mud line. The given rationale was to avoid disturbing the area around the wellhead seal. The unusual depth had an additional advantage: All the very costly mud above the top plug could be pumped to a workboat for transport to another BP project, instead of being left behind.

  But not until the well was completely sealed.

  Ever since the well had penetrated into the oil and gas deposit, the weight of all that mud had been keeping the hydrocarbons from shooting to the surface. Once the drilling mud was replaced with much lighter seawater, control of the oil and gas would depend on the integrity of the well’s seals and plugs.

  Dewey outlined the final test they would perform to ensure that Macondo was safe, and ready for mud displacement. The procedure was called a “negative test.” In the “positive test” that had been run that morning, pressure had been inserted inside the well casing to see if anything leaked out. The negative test would take pressure off the well, to see if anything leaked in. The test was somewhat complicated and would take some time.

  The consumption of time was apparently a sticking point. The first plan that BP company man Bob Kaluza presented for the day consisted of displacing the mud in the riser with seawater and setting the top plug. A negative test was not in the plan.

  Jimmy and Randy had some words with Bob. Skipping the negative test “is not my policy,” Jimmy said. He alluded to a bad experience he’d had years ago, but he didn’t go into detail. “It taught me a lesson,” he said. On the Horizon, Jimmy insisted, a negative test was standard until he heard differently.

  Well, I’m the company man, Kaluza told him in effect, and you’re hearing it from me. If the well passed the positive test, he said, the negative test was overkill.

  Jimmy was adamant. He’d given in on the nitrogen foamed cement. He wasn’t giving in this time. They were running that test.

  The helicopter carrying the VIPs arrived on the rig at 2:30. It was one of those wave-the-flag, check-in-with-the-little-people management exercises. Two senior execs from Transocean, Daun Winslow and Buddy Trahan, had joined up with two BP counterparts, David Sims and Pat O’Bryan. They met in Houma and got lunch together before the 12:30 check-in for the helicopter shuttle out to the rig.

  Daun was the ranking Transocean exec, officially the operations manager, performance, for the North American Division. Like many top Transocean officials he’d come up from the rig floor—he’d been an OIM for a decade, then stepped up to be a shore-based rig manager, the OIM’s boss, and now he oversaw a vast swath of Transocean’s operations. His job description was to oversee “safe and efficient operation, maintenance and equipment on rigs in Gulf of Mexico”—all without a four-year college degree to his name. His highest academic attainment was technical college and an auto body apprenticeship. He had a slightly unkempt look, thinning brown hair swept back like it had frozen in a strong wind. His lean build and smoker’s pallor was underlined by physical confidence that conveyed the habit of command.

  Buddy Trahan, forty-three, was the rig veteran who had flown in to help set the Horizon right after its near capsizing in 2008. He’d come from a large family that had run shrimp boats in the Gulf, and now he s
upervised six Gulf oil rigs worth more than an entire fleet of shrimp boats.

  Together, Daun and Buddy hoped to use the visit to enhance the relationship with their biggest client, wave the flag, and show O’Bryan, BP’s vice president for drilling and completions, and Sims, the company’s drilling operations manager, the in-the-field spirit of one of their most exemplary rigs.

  These VIP trips had another purpose as well. The people who ran both BP and Transocean understood that an us-versus-them dynamic would inevitably arise between the workers on the rigs and the folks on “the beach,” as the Houston headquarters were called. Mostly, the rig tours were an attempt to close the gap, at least a little. For those making the big decisions it was important not to lose touch, to see with their own eyes and hear with their own ears how their policies translated into the tough and dangerous work of pulling oil from the Gulf.

  Even so, it wasn’t in them to pass up a chance to proselytize. They went over a few points they wanted to address—the hazard of dropped objects, of course. Also, and they wrote this down, “Slips, Trips, Falls.”

  They saw a chance to reinforce the values that had served the Horizon so well. The “No blame, ‘can do’ culture—fix the problem, learn, move on.” Also, “prudent risk-taking—freedom to fail, no fear of second guessing.”

  The intense concentration of Transocean and BP on relatively minor “slips, trips, and falls” had struck some as odd. They were, after all, sitting directly above a cavern of highly pressurized, highly flammable material that could erupt, with horrific consequences. But few questioned the motives of executives who were, after all, looking out for their safety.