Pipeline that blew up installed in 1960s
CHARLESTON, W.Va. -- The NiSource natural gas pipeline that blew up in the Sissonville area last week was constructed in 1967, according to new federal records that contradict the company's earlier statements to local officials.
Local officials had said NiSource told them initially that the pipeline dated back to only the 1990s, in which case it would have been subject to more stringent federal requirements.
However, a U.S. Department of Transportation order, dated Thursday, says the "affected pipeline at the failure location was constructed in 1967."
The finding supports new comments from Kanawha County officials, who said NiSource gave them updated information this week revealing that the line dated back to the 1960s.
"It was an older pipeline," said C.W. Sigman, the county's deputy emergency services director.
Conflicting initial reports about the pipeline's age would not have come as a surprise to federal officials, who have found similar problems in previous investigations. And even in the early stages of their investigation of the Dec. 11 blast, officials are seeing similarities to other gas pipeline incidents around the country.
In an interview earlier this week, National Transportation Safety Board Chairwoman Deborah Hersman said initial findings -- including the lack of automatic or remote shutoff valves on the line -- illustrate broader problems facing the nation's 2.5 million miles of pipelines.
"I think we've seem some common themes in Sissonville that we've seen in other accidents," said Hersman, who was a staffer for former West Virginia congressman Bob Wise. "We have seen the same things over and over again in our investigations around the country."
Hersman said the NiSource explosion was a near-miss accident that easily could have caused "fatalities and a number of serious injuries."
"What we've found -- the extent of the corrosion that our investigators found when they were on the scene -- was alarming," Hersman said.
A 10-member team of NTSB investigators found that the pipe that blew up was only 0.078 inches thick -- about 30 percent as thick as it should have been.
"We do not expect to see that type of wall loss in a pipeline that is operating in a high pressure, carrying hazardous materials," Hersman said. "We want to understand how the pipe got to that level of corrosion and why it was not identified."
The NTSB has said the blast occurred on a 20-inch-diameter natural gas transmission line. Several people received minor injuries, several homes were destroyed and the ensuing fire engulfed and damaged a large section of Interstate 77 north of Charleston.
Officials from NiSource subsidiary Columbia Gas Transmission have generally refused to answer questions about the incident.
Columbia CEO Jimmy Staton did issue an "open letter" saying that "something went terribly wrong" at the pipeline and that his company would work with the NTSB to find out what happened and "take every step necessary to ensure the continued safety of our pipeline system."
On Friday, new information about the age of the pipeline involved in the blast surfaced, with the public release of an order from the Transportation Department's Pipeline and Hazardous Materials Safety Administration, or PHMSA, outlining required repairs before the pipeline can be used again.
The PHMSA order said the line in question -- known as SM-80 -- was originally built in 1951 from Lanham to Broad Run, and was extended in 1955 to Leach.
"Since that time, various segments of the pipeline have been replaced, resulting in a line with various vintages of pipe, the newest of which stems from a 1992 project," the order said. "The affected pipeline pipe at the failure location was constructed in 1967."
During a media briefing the day after the explosion, NTSB member Robert Sumwalt declined to answer when asked what the agency's investigators had learned so far about the age and specifications of the pipe that exploded.
"To tell you the truth, it's not as easy of a process as you may think that you're going to know exactly what that pipe is and the attributes of that pipe," Sumwalt said. "We have seen in other pipeline explosions where we go and we're told that the pipe is one thing and we find out it's another.
"I don't want to stand here and tell you what we've heard," Sumwalt said. "I want to make sure that, before I release that information, that we actually have a great deal of confidence that's what it is."
On Friday, Peter Knudson, a spokesman for the NTSB, circulated the PHMSA order to reporters covering the Sissonville explosion but would not say if the pipeline data contained in it matched what the NTSB has found so far.
"Our investigators are still in the processes of confirming the age of the segment of pipeline that ruptured," Knudson said. "We expect to include that determination in the preliminary report."
The preliminary report is due out 30 days after the initial incident, or by Jan. 11.
"Given the volume of information that we request from all the entities involved in a major accident," Knudson said, "it is not at all unusual for there to be some pieces of information that will be refined or corrected as the investigation progresses."
Following a September 2010 natural gas transmission pipeline explosion that killed eight people in San Bruno, Calif., pipeline operator Pacific Gas and Electric told the NTSB that the pipe involved was seamless. NTSB investigators later learned that it was made up of six short segments, known as "pups," that were welded together.
In a 153-page report on the San Bruno disaster, the NTSB found that the explosion was caused by a fracture of one of those six pipe sections, and that fabrication of the sections in 1956 "would not have met generally accepted industry quality control and welding standards then in effect, indicating that the standards were either overlooked or ignored."
NTSB investigators found a variety of problems with PG&E's recordkeeping, and with the database used to track pipeline specifications for purposes of determining the integrity of the system and the need for maintenance or replacements.
"There is a very high and very clear expectation from the regulator that the operator knows their lines, knows the risk and appropriately assesses and deals with those risks," Hersman said. "What we see in accidents is that that hasn't happened."
In the interview Wednesday, Hersman said the NTSB is especially concerned about pipelines put into service before 1970, which are "grandfathered in" and don't have to be hydrostatically tested and often can't be examined with in-line integrity devices, known as "pigs." About half the nation's pipelines fall into the pre-1970 category, she said.
"There is a very real safety concern with pipes that are older and have been subject to less-stringent requirements," Hersman said.
The NTSB has recommended eliminating the grandfather provision for hydrostatic testing and requiring that all lines be in-line tested, but those measures were not part of a pipeline safety bill passed by Congress last year and signed into law by President Obama in January.
The PHMSA order contains a few other bits of information, including times for the explosion and the gas flow shutdown, that conflict slightly from what the NTSB previously reported.
The NTSB had said the explosion occurred at 12:41 p.m., and that NiSource crews were not able to manually shut off the flow of gas to the fire until 1:45 p.m. The PHMSA order says the explosion occurred at 12:43 p.m., and that one end of the pipeline was shut off manually at 1:20 p.m. and the other at 1:40 p.m.
NTSB officials have said conflicting times are common in investigations of such incidents, because computers and clocks in various locations aren't synchronized.
The PHMSA report said the pipeline failure "resulted in the release and ignition of an undetermined amount of gas and created two flame plumes." It said 3 homes were destroyed, another seriously damaged and several others damaged.
"We were so fortunate in this accident that there were not fatalities and a number of serious injuries," Hersman said. "From the team that was on the scene, there was a great deal of devastation. We don't want to see this happen again in another community in West Virginia."
Reach Ken Ward Jr. at email@example.com or 304-348-1702.