AD Main Menu

BP says 'F' rankings show monitoring system is working

Abrahm LustgartenProPublica

Here's a little more detail on what it means that 148 of BP's pipelines in Alaska had been ranked for "failure" by BP inspectors, and a more robust explanation from BP than it offered for the main story.

According to a BP report generated on Oct. 1, 351 isolated locations on 148 unique sections of pipes or pipelines have been given an "F" ranking and are deemed at risk of failing.

That includes larger outdoor oil and gas lines as well as smaller waste pipes and indoor drain lines, but all of them can transport hazardous or flammable material.

The F-ranks are based on hot-spots -- corroded locations along the line that can range from less than an inch to a couple of feet in size -- and may be worn as thin as a few thousandths of an inch.

It doesn't take much to cause a spill. When BP's oil transit line -- which was not "F-ranked" -- spilled 212,000 gallons of oil onto the snow in 2006, the North Slope's largest spill, it was found to have seeped through a corroded hole less than half an inch wide.

According to the BP data, some pipelines have many locations worn thin enough to warrant the rating. One waste line for produced water has 42 locations at risk of wearing through.

When such a location is identified, it is given the "F" ranking, the most severe rating in the company's routine maintenance program and an indication that more than 80 percent of the thickness of the pipe in question has corroded.

"It means the pipe wall is corroded to the point where it has to be retired," said Martin Anderson, a former pipeline inspection supervisor for a BP contractor. An F-rank can also be occasionally assigned based on problems besides corrosion, like a dented or cut pipe, he said.

BP told ProPublica that the F-rankings indicate the company's monitoring is working well -- the inspectors are identifying trouble spots so that BP can fix them and prevent a spill. The company emphasized that it maintains over 1,600 miles of pipelines, and it is not unusual to have 148 lines listed. It said there are always "F" ranks working through the system.

"To our knowledge, we've never had a line with a F-rank defect fail in service," said BP Alaska's spokesman, Steve Rinehart.

Rinehart said that as of Nov. 3 the company had 61 pipelines with 151 hot spots "F-ranked." The data is different from ProPublica's, he said, because it relates specifically to the pipeline systems extending from the well pads through the production system, and may not include smaller lines that are ancillary to the process. Of the 151 F-rankings that he described, 15 are on out-of-service pipelines, 17 are on pipelines where the flow or pressure has been reduced, 104 are on operating pipelines with scheduled repairs and 15 have already been repaired or resolved.

But workers like Marc Kovac, the BP employee who provided the data to ProPublica, allege that the company has not been quick enough in addressing problems. Instead, said Kovac, many of the pipelines have been allowed to languish with temporary patches or postponed repairs.

The F-rank data obtained by ProPublica lists inspection dates for when the rankings were assigned and shows that many of the lines' problems date back to 2008 and 2009. A smaller number of lines were inspected in 2005. One is dated 1994.

"An item can be an F-rank and can stay that way for some period of time as long as the situation is stabilized," said Rinehart, who says a pipe can be patched, shut down, or de-rated to handle less material. "It's a big job and we're all over it."

"BP has not responded properly over the years and allowed the problems to pile up," Kovac said. "It's a joke in the field to say 'this is a temporary permanent patch.' We know the patch will be in place for many years."

The document showing the "F-rank" corrosion information consists of data Kovac copied from its original form into this sheet, in order to protect the identity of the worker who first ran the report. ProPublica verified the data against photographs of the original print-outs, and shared a sample of the data with BP prior to writing about it.


By ABRAHM LUSTGARTEN
ProPublica