Good morning Madame Chairwoman and Members of the Committee. It is a pleasure to appear today to discuss safety issues related to the transportation of hazardous materials.

The Safety Board has been concerned over the transportation of hazardous materials in all modes of transportation for many years. Over the past 20 years, however, we have seen many improvements made to enhance the safe transportation of hazardous materials. Some of those improvements include:

! enhanced packaging requirements, such as head shields and top and bottom shelf couplers on railroad tank cars carrying the most dangerous cargoes to reduce the likelihood of head punctures and loss of product, and thermal protection to allow time to evacuate areas around accident sites before tanks violently rupture when those tank cars are involved in fires;

! improved inspection and repair requirements for highway cargo tanks; and

! improved requirements for communicating information about the hazards posed by cargoes transported, including better shipping descriptions, 24-hour telephone numbers for contacting shippers, initial information about potential hazards, public safety, and emergency response actions.

I would like to discuss, however, four areas where the Board believes that additional action should be taken to enhance the safe transportation of hazardous materials. Those areas include: testing requirements for tank cars to identify structural defects before sudden and complete failure; rollover protection requirements for highway cargo tanks; crashworthiness of front heads on cargo tanks used to transport liquefied flammable gases and potentially lethal nonflammable gases; and practices, procedures, and training to identify undeclared hazardous materials shipments.

On January 18, 1992, in Dragon, Mississippi, a railroad tank car carrying more than 30,000 gallons of liquefied propane literally pulled apart as a train began to move from a siding onto a main track. All 30,000 gallons of propane were instantaneously released. Following ignition of propane gases, the entire area was engulfed in fire. The Safety Board=s investigation found that a crack that had initiated on the inside of the tank had grown to 21 inches long before the tank catastrophically failed. Post-accident inspection and testing of 108 tank cars of the same design disclosed that 40 of these tank cars had cracks in the same location.

A second accident occurred on March 25, 1992, in Kettle Falls, Washington, involving a railroad tank car carrying 13,000 gallons of sulfuric acid. The tank car cracked at the bottom center of the tank along a circumferential weld, resulting in the release of all the sulfuric acid. There was metallurgical evidence of a preexisting crack in the area of the failure. The tank car had just passed visual inspections and a hydrostatic test the previous month, and it was carrying its first load of cargo since the inspection and test. The tank car failed as the train began to move forward.

Because of these structural tank car failures, the Safety Board conducted a special investigation on the inspection and testing of railroad tank cars. The Board concluded that hydrostatic tests and visual inspections of tank cars at arbitrary intervals were not effective in detecting structural defects in welded tank car tanks or in detecting defects at high stress areas where stub sills or other components are attached to tanks before sudden and complete failure.

The Board found that the use of acoustic emission, ultrasonic, and other nondestructive testing methods, if applied at appropriate intervals based on damage-tolerance concepts, could detect existing cracks prior to catastrophic failure of rail tank cars. However, additional research was needed to determine the best methods to be used under given conditions and the appropriate inspection intervals. The Safety Board recommended that the Federal Railroad Administration (FRA), the Research and Special Programs Administration (RSPA), the Association of American Railroads (AAR), the Railway Progress Institute and the Chlorine Institute work together to evaluate non-destructive techniques and determine how such techniques can best be applied for periodic testing and inspection of all tank cars that transport hazardous materials, and that the FRA and RSPA then develop requirements for the periodic inspection and testing of tank cars that help ensure the detection of cracks before they propagate to critical length.

We are pleased that the FRA, RSPA and the industry have agreed with recommendations to evaluate ways to improve inspection and test techniques for tank cars and are co-funding several research projects. On September 21, 1995, the RSPA and FRA published two final rules that will improve the reliability of detecting cracks, pits, corrosion, and lining flaws in tank cars by the use of nondestructive tests, and to establish base tank car inspection intervals acquired through damage tolerance analysis. Full implementation of a damage tolerance program will take several years. The Board has asked the FRA for periodic updates on its progress in implementing the recommendation. (Safety recommendation R-92-22 remains AOpen--Acceptable Action.)

Between January and May 1991, the Safety Board investigated 7 highway accidents in which DOT specification MC 306 or MC 312 cargo tanks overturned and hazardous materials were released through damaged closures or fittings on top of the tanks. The highway cargo tanks were equipped with rollover protection devices. However, the failure of the devices to provide adequate rollover protection raised concerns about their performance, and the adequacy of DOT requirements regarding the structural integrity and configuration of the rollover protection devices.

As a result of its special investigation on cargo tank rollover protection, the Safety Board found deficiencies in cargo tank rollover protection requirements. On March 20, 1992, the Safety Board recommended that the Federal Highway Administration (FHWA), with the assistance of RSPA, improve the performance of rollover protection devices on bulk liquid cargo tanks by modeling and analyzing the forces that can act upon these devices during a rollover accident, and then promulgating performance standards for rollover protection devices that are based on the engineering modeling and analysis.

In its most recent response to the recommendations, on April 10, 1996, the FHWA advised that it was evaluating cargo tank crash modeling, including the use of computer-based techniques and finite element analysis to simulate accident forces and to correlate the computer modeling with real world accident situations. The Safety Board is concerned about progress on this issue and has asked FHWA for additional information about this model.

The Safety Board is also concerned about the crashworthiness of front heads on cargo tanks used to transport liquefied flammable gases and potentially lethal nonflammable compressed gases. On July 27, 1994, in White Plains, New York, a tractor cargo-tank semitrailer loaded with 9,200 gallons of propane hit a bridge column, fracturing the front head and releasing propane. When ignited, a fireball rose 200 to 300 feet in the air. The tank was propelled northward about 300 feet and landed on a frame house, engulfing it in flames. The driver was killed and 23 people were injured.

Following a derailment in Crestview, Florida in 1979 that resulted in the failure of several rail tank cars carrying liquefied compressed gases, the Safety Board issued an intermodal safety recommendation to RSPA asking that it expand its research into new approaches for controlling pressurized liquefied flammable gas releases from breached tanks. (I-79-12) In 1991, 12 years after issuance of our safety recommendation, RSPA advised the Safety Board that the research program had been canceled, but that research had not yielded any viable alternatives to railroad shelf-couplers, headshields, and thermal protection, all which had proven effective in preventing product release. The Safety Board agreed that rail tank car improvements had reduced the number of catastrophic failures of pressurized tank cars, but reminded RSPA that the safety recommendation was intermodal and that RSPA had not addressed this problem as it related to highway cargo tanks.

To further support research, the Safety Board advised RSPA about 4 highway accidents it had investigated involving the failure of cargo tanks carrying LPG. Those accidents occurred April 29, 1975, at Eagle Pass, Texas, resulting in 51 burn injuries and 16 fatalities; April 6, 1987, in Lawrenceville, New Jersey, resulting in 7 burn injuries; December 23, 1988, in Memphis, Tennessee, resulting in 23 burn injuries, 9 fatalities; and January 20, 1992, in Crawford, Mississippi, resulting in 4 burn injuries and 3 fatalities. RSPA did not respond, and on June 29, 1994, the Safety Recommendation I-79-12 was AClosed--Unacceptable Action.@

Because of the severe accident in White Plaines, New York, the Safety Board again called for action to improve the crashworthiness of cargo tanks. On November 27, 1995, the Safety Board asked FHWA to cooperate with RSPA in studying methods and developing standards to improve the crashworthiness of front heads on cargo tanks used to transport liquefied flammable gases and potentially lethal nonflammable compressed gases (H-95-35). In responding to the safety recommendation on April 10, 1996, the FHWA agreed with the Safety Board that the crashworthiness of the front end of cargo tanks used to transport these cargoes needed further evaluation. It stated that it would request funds for this project during fiscal year 1997. On June 24, 1996, Safety Recommendation H-95-35 was classified AOpenCAcceptable Response.@

Finally, Valujet flight 592 crashed on May 11, 1996, shortly after takeoff from Miami International Airport, Florida. Both pilots, the three flight attendants, and all 105 passengers were killed. Before the accident, the flightcrew reported to air traffic that it was experiencing smoke in the cabin and cockpit.

Factual information provides that five cardboard boxes containing as many as 144 chemical oxygen generators, some with unexpended oxidizer cores, and three wheel/tire assemblies, had been loaded in the forward cargo compartment of flight 592 shortly before departure. These items were being shipped as company material (COMAT). Additionally, some passenger baggage and U.S. mail were loaded into the forward cargo compartment.

Shortly before the departure of flight 592, a driver from the SabreTech Inc., maintenance facility at the Miami airport delivered the COMAT (the boxes and wheel/tire assemblies) to the ValuJet lead ramp agent for transport to ValuJet facilities in Atlanta. (SabreTech operated an FAA-approved aircraft repair and maintenance facility at the Miami airport and had performed renovation work for ValuJet.) A SabreTech shipping ticket, dated May 10, 1996, for the five boxes of chemical oxygen generators, was also offered to the ramp agent. The generators were identified on the shipping ticket as AOxy Canisters [sic] AEmpty@.@

The ramp agent, who was busy offloading the aircraft from its previous flight, signed the shipping ticket for the COMAT and instructed the SabreTech driver to place the items on an empty baggage cart. The ramp agent stated that he asked the first officer of flight 592 for approval to load the COMAT on the aircraft. After the ramp agent and the first officer estimated the weight of the COMAT, the baggage and mail were loaded into the cargo compartment, and then the three wheel/tire assemblies were loaded, followed by the five boxes with the generators. The boxes with the generators were loaded on top of or immediately adjacent to the wheel/tire assemblies.

Chemical oxygen generators, when transported as cargo, are considered a hazardous material regulated under the Department of Transportation hazardous materials regulations (49 CFR Parts 171-180) and are classified as oxidizers. These same regulations govern the packaging, labeling, and shipping requirements for transportation of chemical oxygen generators.

Although the origin of the in-flight fire on board flight 592 has not yet been determined, the presence of the chemical oxygen generators in the forward cargo compartment of the aircraft created an extremely dangerous condition. The chemical decomposition reaction of an oxidizer such as sodium chlorate in a confined space will generate heat, and the oxygen resulting from the reaction will sustain and intensify a fire. Also, the ignition temperature of ordinary materials is lowered in an oxygen-rich environment.

The failure to properly identify and properly package hazardous materials had resulted in other accidents and incidents.

For example, on November 3, 1973, a Pan American World Airways, Inc., Boeing 707-321C crashed at Logan International Airport, Boston, Massachusetts, killing all three crewmembers. Thirty minutes after this cargo flight departed John F. Kennedy Airport, New York, the flightcrew reported smoke in the cockpit, and the flight was diverted to Logan, where it crashed short of the runway. The Safety Board determined that dense smoke in the cockpit seriously impaired the flightcrew=s vision and ability to function effectively during the emergency. Although the source of the smoke could not be established conclusively, the Board believed that spontaneous chemical reaction between leaking nitric acid (a corrosive and oxidizing material), which was improperly packaged and stowed, and the improper sawdust packing surrounding the acid=s package, initiated the accident sequence. A contributing factor was found to be a general lack of compliance with existing regulations governing the transportation of hazardous materials and the inadequacy of government surveillance. Further, the Safety Board concluded that most personnel handling the hazardous material shipment were inadequately trained.

On August 10, 1986, a McDonnell Douglas DC-10-40, operating as a nonscheduled flight from Honolulu, Hawaii, to Chicago, Illinois, with an en route stop in Los Angeles, California, arrived without incident at Chicago=s O=Hare International Airport. After the passengers and crew had deplaned, a fire, which was found to have initiated in a cargo compartment, burned through the cabin floor, spread rapidly throughout the entire cabin, and destroyed the airplane.
 

The Safety Board concluded that the fire had been initiated as a result of a mechanic=s improper handling of a chemical oxygen generator associated with a seatback temporarily stored in the compartment. The Safety Board learned as a consequence of this incident that some air carriers were not aware that solid-state passenger supplemental oxygen generators were capable of generating high temperatures and were classified as hazardous materials when carried as company material in cargo compartments. Consequently, some air carriers were not taking the required precautions when shipping oxygen generators in their airplanes. Following this incident, the FAA promptly notified all domestic air carriers and foreign airworthiness authorities of the circumstances of the incident and reminded them that oxygen generators are oxidizers and therefore are classified as hazardous materials, which should be packaged and stowed securely.

On February 3, 1988, American Airlines flight 132, a DC-9-83, had an in-flight fire while en route to Nashville Metropolitan Airport, Tennessee, from Dallas/Fort Worth International Airport, Texas. As the aircraft was on a final instrument landing system approach, a flight attendant and a deadheading first officer notified the cockpit crew of smoke in the passenger cabin. The Safety Board found that hydrogen peroxide solution (an oxidizer) and a sodium orthosilicate-based mixture had been shipped and loaded into the mid cargo compartment of the airplane. The shipment was improperly packaged, and it was not identified as a hazardous material. After the hydrogen peroxide leaked from its container, a fire started in the cargo compartment. The fire eventually breached the cargo compartment, and the passenger cabin floor over the midcargo compartment became hot and soft. Fortunately, the aircraft landed without further incident, and the 120 passengers and six crewmembers safely evacuated the aircraft.

As a result of the accident on American Airlines flight 132, the Safety Board stated that in addition to proper packaging of hazardous materials, the safe transportation of hazardous materials depends on sufficient information to identify the materials and the hazards presented during transportation. Accordingly, the Board noted that both shippers and carriers had a responsibility to determine if materials offered for transportation were hazardous and in proper condition to ensure their safe transportation.

The Safety Board stressed in its report of the American Airlines incident the importance for air carriers to have effective policies, practices, and training to screen passenger baggage and freight shipments for undeclared or unauthorized hazardous materials that are offered for transport. However, acceptance of undeclared and unauthorized shipments of hazardous materials continues to pose a significant threat to passenger and cargo aircraft. Consequently, the Safety Board believes that the FAA should evaluate the practices of all air carriers, including training, for accepting passenger baggage and freight shipments (including company material) and for identifying undeclared or unauthorized shipments of hazardous materials, and then require all air carriers to revise their procedures as necessary.

The Safety Board has yet to complete action on the ValuJet accident that occurred near Miami. However, as a result of our investigation, the Safety Board issued a safety recommendation letter to the Federal Aviation Administration (FAA) that called for urgent actions to prevent similar accidents in the future. Those recommendations state:

Immediately evaluate the practices of and training provided by all air carriers for accepting passenger baggage and freight shipments (including company materials) and for identifying undeclared or unauthorized hazardous materials that are offered for transport. This evaluation should apply to any person, including ramp personnel, who accepts baggage or cargo for transport on passenger and cargo aircraft. (A-96-25)

Require air carriers, based on the evaluation performed under Safety Recommendation A-96-25, to revise their practices and training for accepting passenger baggage and freight shipments and for identifying undeclared or unauthorized hazardous materials that are offered for transport. (A-96-26)

In a response dated July 16, 1996, the FAA agreed with the recommendations and stated that it had initiated an immediate evaluation of air carrier hazardous materials manuals and training programs to determine if methods are included to identify undeclared or unauthorized hazardous materials. On March 24, 1997, the FAA provided an update on its response to these recommendations. It noted that the FAA had completed its first phase of its evaluation and had reviewed 215 air carriers certified under 14 CFR Parts 121, 125, and 135. During the evaluation, the FAA said it reviewed FAA-approved air carrier manuals to ascertain what information is currently provided to prompt air carrier employees to recognize suspicious cargo or baggage and to ask additional questions before rejecting or accepting that cargo or baggage. The FAA said that in it had initiated phase 2 of the evaluation which consists of followup field G5
inspections of those carriers to verify that the air carriers which have procedures in their manual are following those procedures. Based on the outcome of phase 2, the FAA noted that it would have all the necessary information to recommend appropriate actions to correct any noted deficiencies and to address the issues outlined in the safety recommendations.

The Safety Board is concerned, based on the facts developed during the ValuJet flight 592 accident investigation, that the practices, procedures, and training of the personnel involved in the identification and handling of hazardous materials remain inadequate.

Even more so, Madame Chairwoman, we are concerned that, although the FAA has banned oxygen generators from passenger airliners and has proposed a ban on other oxidizers, no action has yet been taken on recommendations we issued nine years ago to install smoke detectors and fire suppression equipment in cargo compartments that do not currently have them.

Those recommendations, issued after the Nashville accident, were closed by the Board and classified "Unacceptable Action." Last November, the airline industry announced it would voluntarily retrofit existing Class D compartments with smoke detectors. We are watching carefully to see if this retrofitting will take place in a timely fashion.

The fact still remains, Madame Chairwoman, that pilots taking off today in aircraft having Class D cargo compartments have no more ability to be alerted of the existence of a fire in that compartment, much less the ability to suppress it, than did the crew of ValuJet flight 592, or indeed, the crew of the American Airlines airplane in 1988.

Madame Chairwoman, that completes my statement, and I will be pleased to respond to any questions you or the Committee may have.