Exxon: Can't settle on cause of rupture

Consultant says that’s hard to believe

Exxon Mobil has told federal regulators that it could not determine what sped the growth of decades-old cracks in its Pegasus pipeline, causing the line to rupture and spill crude oil into a Mayflower neighborhood last year.

A pipeline safety consultant advising Central Arkansas Water countered Monday that he doesn't believe the oil giant's statement.

"It's just very odd that they couldn't ascertain a reason ... with some degree of confidence," said consultant Richard Kuprewicz, who serves on a technical advisory committee to the federal Pipeline and Hazardous Materials Safety Administration, often called PHMSA.

Asked if he believed Exxon Mobil on this point, he replied, "No," and said he was "very disappointed."

The roughly 850-mile-long Pegasus pipeline has been idle since it cracked open March 29, 2013, between two houses in Mayflower's Northwoods subdivision and spilled an estimated 210,000 gallons of thick crude into the neighborhood, drainage ditches and a cove of Lake Conway.

Almost a year later, on March 28, Exxon Mobil took a major step toward restarting the line's northern section by filing a proposed remedial work plan with the federal regulatory agency. But neither the company nor the safety administration had released anything more than a summary of the proposal until last week, when the agency posted it online.

In the seven-page remedial work plan filed March 28, Exxon Mobil Pipeline Co. noted that a Texas laboratory had found manufacturing defects -- specifically hook cracks -- as the cause of the rupture in the pipeline, built in 1947-48.

But the company added, "The degradation mechanism of the hook crack defect to failure was undetermined."

Exxon Mobil said in the report that the laboratory had "found no evidence" of equipment failure, internal or external corrosion, stress-related corrosion cracking, or welding or fabrication-related defects. Also not detected were third-party damage, weather-related or outside-force damage, and pressure-cycle-induced damage, Exxon Mobil said.

Since the laboratory report was released in July 2013, Exxon Mobil said, additional analysis has eliminated the possibilities of operator error or operating pressure as factors. The pipeline was functioning well below the maximum allowed pressure when the rupture occurred.

The company listed five factors that it said could cause cracks to grow faster and said the Mayflower accident could have resulted from a combination of them.

Those factors included the pipe's toughness; residual stress within the pipe or its seam; contribution from adjacent defects; pipe fatigue caused by the pressure cycle, although the cycling was "light;" and environmentally induced, or hydrogen-stress, cracking.

Exxon Mobil had never before said publicly that it couldn't determine what specifically sped up the cracks' growth.

Kuprewicz said there are reasons that Exxon Mobil might not want anyone to know that information.

"The question is ... is this something that is related to one little piece of pipe, or are there issues related to the rest" of the pipeline, he said. "While it may or may not be an isolated thing, you just can't conclude it's related just to this piece of pipe" without knowing more.

It's "very unusual" for an oil company not to determine this factor, Kuprewicz said.

As a result, he said, his advice to Central Arkansas Water is that "you've got to assume that this pipe has got problems" beyond Mayflower.

Central Arkansas Water officials have been concerned because about 13.5 miles of the Pegasus travel through the Lake Maumelle watershed. The lake provides drinking water to about 400,000 Arkansans.

The remedial work plan is subject to the safety administration's approval.

Agency spokesman Damon Hill said Monday that the administration has not yet acted on the plan. It also has not ruled on whether the oil company violated nine federal safety regulations and should be fined $2.66 million, as proposed, he said.

The company has submitted a separate work plan for the 211-mile southernmost portion of the pipeline, which extends from the Gulf Coast of Texas to Patoka, Ill. The agency approved that plan in March, and Exxon Mobil spokesman Aaron Stryk has said the company hopes to restart the southern portion this month.

Stryk said Monday that the accident investigation points to "atypical pipe properties as the most significant contributing factor that led the original manufacturing defects to grow to failure."

He added, "The remedial work plan specifically indicates that 'the combination of the manufacturing defects and atypical pipe properties rendered the pipe seam susceptible to many different crack growth mechanisms, resulting in relatively unpredictable crack growth rates.' The plan then listed the potential mechanisms that could have been at play.

"It is not necessary to pinpoint a single mechanism or the exact percentage of each mechanism's contribution to the failure to develop a remedial work plan," Stryk wrote in an email. "Indeed, the plan developed was designed to mitigate the threats associated with all of the potential mechanisms."

Stryk said the company still expects the remedial process to take more than a year once the safety administration approves the work plan.

The Arkansas attorney general's office, which along with the U.S. Justice Department has sued two of Exxon Mobil Corp.'s subsidiaries, "does not have the technical expertise to evaluate" the remedial work plan, Aaron Sadler, a spokesman for Attorney General Dustin McDaniel, said Monday. "But we would expect federal regulators at PHMSA to make sure it addresses the requirements of its corrective action order."

Exxon Mobil proposes to assess the northern portion of the Pegasus through hydrostatic, or water, pressure testing, the remedial plan says.

Those tests are to follow excavation, examination, evaluation and repair when needed of any anomalies previously identified. The pipe also will be examined to determine if excavated segments exhibit similar properties to the section that failed in Mayflower, the plan says.

It adds: "If a significant number of pressure-reversal failures occur, [the pipeline company] may decide to reduce the targeted test pressures in order to complete the testing in more efficient manner. Should this become necessary, the resulting MOPs [maximum operating pressures] will be at the same ratio of test pressure to MOP, i.e. MOP will not be more than 72 percent of the peak spike test pressures obtained in each section."

Kuprewicz questioned the wisdom of this plan.

"The whole purpose" of a hydrostatic test is to find any vulnerable areas. It's far better for a pipe carrying water to crack open than one carrying oil, he said. "If you have failures ... and just start lowering the pressure," Kuprewicz said, "that is not what you [should] do."

State Desk on 07/15/2014

Upcoming Events