(The following article by Larry Bingham was posted on the Oregonian website on April 5.)
STEVENSON, Wash. — Railroad crews and track inspectors reported multiple problems with a curved stretch of track in southern Washington just days before an Amtrak train bound for Portland derailed there Sunday morning, injuring 30.
Cy Gura, an investigator with the National Transportation Safety Board, cited four reports of track abnormalities and said 18 concrete ties had worn down on the track, causing the Amtrak locomotive and its cars to jump off the rails near Home Valley, Wash. Twenty-four of the train’s 105 passengers and six of its nine crew members were treated at hospitals in Washington and Oregon.
Why the track problem wasn’t discovered sooner will be the target of an NTSB investigation.
“We have a good idea of what happened,” Gura said Monday. “Now we have to figure out why it happened.”
The four reports are accompanied by a September 2004 report that cites track abnormalities around the site of the derailment, Gura said. None of the reports specifically cited curve 58-B, where Sunday’s derailment occurred.
The first report came Sept. 23, 2004, when a track geometry vehicle flagged 10 sites in yellow, including curve 58-B, along a 10-mile stretch near the crash site. The yellow flag citation meant structural elements of the track should be measured and monitored frequently, Gura explained.
A yellow flag citation is one degree below a red flag, requiring immediate repairs. But the four reports will be at the core of the NTSB’s investigation.
The first of the four came March 23, when a federal track inspector reported a “rough ride” at milepost 58.4, about 250 feet from where the train derailed, Gura said.
Investigators have found the report was radioed to Burlington Northern Santa Fe Railway Co., which owns the tracks. But investigators do not have a record of Burlington track inspectors checking the problem, other than an e-mail sent to the track inspector saying Burlington would look into the report, Gura said.
Burlington Northern spokesman Gus Melonas on Monday referred questions to the NTSB.
The second report of a rough ride came five days later, on March 28, when an Amtrak crew reported an abnormality in the same general area, Gura said.
“If you have four reports in a short period of time, and they’re constant, that would be a place to look,” Gura said Monday.
After the second report, investigators learned the speed along the area was reduced to 20 mph, and a Burlington track inspector went to the site to check the track. The inspector, Gura said, thought he discovered a “deflection on the bridge approach” as potentially triggering the rough ride report. But after measuring the approach, he found it within acceptable tolerances, and the speed along the track was returned to 60 mph.
Another Amtrak crew March 30 reported a rough ride in the same area, Gura said. Investigators have not interviewed the Burlington track inspector who checked that report, but hope to today.
The fourth report of a rough ride came April 1, two days before the derailment. A Burlington Northern crew member radioed in that report, but investigators have not yet listened to the taped recording to learn the specifics of his concerns, Gura said.
The NTSB’s full investigation could take five months or longer, Gura said.
“We’re looking at the inspections process, the reporting of the problem, how that is inspected and repaired,” he said.
Meanwhile, new track was laid following the removal of damaged rail cars, and the rail line reopened Monday.