(The following story by Dave Caldwell appeared on the Minot Daily News website on January 18, 2009.)
MINOT, N.D. — The effects on the railroad industry of catastrophic rail accidents such as the Minot derailment of 2002 still reverberate.
The Minot derailment played a major role in the National Rail Safety Action Plan, which was launched in May 2005 in response to that and several other additional major accidents. That plan proved to be an indication that the Federal Railroad Administration was addressing critical issues in an aggressive manner in an attempt to mitigate effects of subsequent derailments or other rail accidents.
Ironically, according to the NRSAP, 2002 was the only year over a six-year period that saw a decline in accidents from the previous year. Between 1998-2004, accidents increased an average of 5.3 percent per year except for 2002, when the number actually fell 9.4 percent. Train accidents occurred an average of 3.8 times per million train miles in 2002. The number did not fall that low again until 2006, after the enactment of the NRSAP.
Among the focuses of that plan were the enhancement of hazardous materials safety and emergency preparedness, better focusing of the FRA’s inspection and enforcement actions on the gravest areas of safety concern, and improving track safety.
The derailment occurred when a Canadian Pacific Railway freight train traveling east at 41 mph on the Portal Subdivision derailed 31 of its 112 cars about one-half mile west of Minot, near the Tierracita Vallejo area. Fifteen tank cars carrying anhydrous ammonia, a pungent gas used most often as a fertilizer, derailed. Five of the cars ruptured catastrophically, releasing 146,700 gallons of anhydrous instantaneously. Another 74,000 gallons leaked from six additional cars over the course of the next five days.
The accident occurred at 1:37 a.m. The temperature at that time was around minus 5 degrees, and a temperature inversion was occurring, which acted to keep the cooler air closer to the ground and keeping the anhydrous from dissipating quickly. There was also only slight wind that night, coming from the southwest, which allowed the plume of poisonous gas, which rose to an estimated height of 300 feet, to blow straight toward downtown Minot and work its way slowly through the area.
One man, John Grabinger, 38, died after being overcome by the fumes while trying to flee his home. Hundreds more were injured, some seriously, including the train crew members and many emergency responders.
One of the derailed tankers ruptured so violently that a piece flew 1,200 feet and struck a house. Two people were sleeping in the room where the piece struck. Thirty of the cars were completely destroyed and the 31st damaged, with a loss of nearly $2.5 million to the railroad. About 475 feet of track was also destroyed.
The cause of the derailment was determined to be a broken rail joint. The accident occurred at or near a 36-foot “plug” that had been inserted between the continuous welded rail. The plug was held in place by 36-inch joint bars, which were secured with bolts to the rails on the inside and outside on either end of the plug.
CP Rail personnel had inspected that section of the track the day before and found no track deficiencies. Common practice used by CP Rail during cold weather was to inspect track from inside a Hy-Rail vehicle. Inspectors would check the track visually and by listening for telltale sounds that would indicate loose or defective joints. FRA regulations at that time required the inspections to take place at least twice per week, with at least one calendar day between the inspections. Records indicated that CP Rail inspected the track four to five times per week.
The FRA reconstructed the rail from the accident site and found that the joint bars showed vertical cracks from fatigue that would have been visible during an on-the-ground inspection. When the joint bars fractured, the rail also fractured, causing the fourth car behind the locomotives to derail. It was determined that the rail fracture had occurred either while the train before or as the accident train passed the fracture.
The FRA inspected the track between Minot and Portal after it reopened, finding seven more cracked joint bars that were not found by CP Rail inspections performed immediately post-accident. An unscientific “drop test” consisting of the dropping of a joint bar across a railhead from a height of five feet found that bars with as little as a one-eighth inch crack would fracture similarly to the joint bar at the site of the derailment. The FRA issued a “special notice of repair” action against CP Rail, temporarily reducing the speed on the Portal and two other subdivisions to 25 mph for failure to comply with proper procedures.
The ineffective inspection and maintenance program’s failure to identify and replace those cracked joint bars was labeled the primary cause of the accident. However, FRA’s oversight of CP Rail’s inspection procedures was also determined to be ineffective. CP Rail had submitted its inspection program as required to the FRA, but three years later, that program had not been compared to the federal standards when the accident occurred.