Skip Ribbon Commands
Skip to main content
Safety Recommendation Details

Safety Recommendation R-83-092
Details
Synopsis: ON SEPTEMBER 28, 1982, ILLINOIS CENTRAL GULF RAILROAD (ICG) FREIGHT TRAIN EXTRA 9629 EAST (GS-2-28) DERAILED 43 CARS ON THE SINGLE MAIN TRACK OF THE HAMMOND DISTRICT IN LIVINGSTON, LOUISIANA. OF THE DERAILED CARS, 36 WERE TANK CARS; 27 OF THESE CARS CONTAINED VARIOUS REGULATED HAZARDOUS OR TOXIC CHEMICAL COMMODITIES, 2 CONTAINED NONREGULATED HAZARDOUS MATERIALS, AND 5 CONTAINED FLAMMABLE PETROLEUM PRODUCTS. A TOTAL OF 20 TANK CARS WERE PUNCTURED OR BREACHED IN THE DERAILMENT. FIRES BROKE OUT IN THE WRECKAGE, AND SMOKE AND TOXIC GASES WERE RELEASED INTO THE ATMOSPHERE. THERMALLY INDUCED EXPLOSIONS OF TWO TANK CARS THAT HAD NOT BEEN PUNCTURED CAUSED THEM TO ROCKET VIOLENTLY. ABOUT 3,000 PERSONS LIVING WITHIN A 5-MILE RADIUS OF THE DERAILMENT SITE WERE EVACUATED FOR AS LONG AS 2 WEEKS. NINETEEN RESIDENCES AND OTHER BUILDINGS IN LIVINGSTON WERE DESTROYED OR SEVERELY DAMAGED. MORE THAN 200,000 GALLONS OF TOXIC CHEMICAL PRODUCT WERE SPILLED AND ABSORBED INTO THE GROUND, REQUIRING EXTENSIVE EXCAVATION OF CONTAMINATED SOIL AND ITS TRANSPORTATION TO A DISTANT DUMP SITE. THIS HAS RESULTED IN LONG-TERM CLOSURE OF THE RAILROAD LINE AND AN ADJACENT HIGHWAY. PROPERTY DAMAGE HAS BEEN ESTIMATED TO BE IN EXCESS OF $14 MILLION.
Recommendation: THE NTSB RECOMMENDS THAT THE CHEMICAL MANUFACTURERS ASSOCIATION: EXTEND THE USE OF COLOR CODING OF TANK CARS OR ADOPT SOME OTHER EFFECTIVE MEANS OF IDENTIFYING HIGH-RISK COMMODITY TANK CARS IN SWITCHING OPERATIONS AND IN WRECK CLEARING OPERATIONS.
Original recommendation transmittal letter: PDF
Overall Status: Closed - Unacceptable Action
Mode: Railroad
Location: Livingston, LA, United States
Is Reiterated: No
Is Hazmat: No
Is NPRM: No
Accident #: DCA82AR016
Accident Reports: Derailment of Illinois Central Gulf Railroad Freight Train Extra 9629 East (GS-2-28) and Release of Hazardous Materials
Report #: RAR-83-05
Accident Date: 9/28/1982
Issue Date: 8/12/1983
Date Closed: 2/18/1987
Addressee(s) and Addressee Status: Chemical Manufacturers Association (Closed - Unacceptable Action)
Keyword(s): Hazmat

Safety Recommendation History
From: Chemical Manufacturers Association
To: NTSB
Date: 5/29/1987
Response: This is to acknowledge receipt of your letter of May 14, 1987, pertaining to NTSB's recommendation R-83-92. While we are disappointed that you have elected to classify our response as "Closed-Unacceptable Action," status, we will continue to pursue the use of transponders fir tank car content identification. The chemical and railroad industries are evaluating the effectiveness of using transponders to provide emergency response personnel with vital information about the contents of individual tank cars. We believe that such devices will prove to be far more effective than tank car color coding. As we develop the specifics of our analysis, we will be pleased to share them with your office. In the meantime, you may wish to contact the Federal Railroad Administration, who we understand has a demonstration project underway dealing with the use of transponders for emergency notification. We would like to pursue this with you further. The next meeting of CMA's Distribution Committee has been scheduled for July 14 at CMA. We invite you to attend the meeting to discuss tank car identification issues with the Distribution Committee. Please have your associate contact Richard Doyle of my staff to let him know if and when it would be convenient for you to participate at our meeting. Mr. Doyle can be reached at 887-1272.

From: NTSB
To: Chemical Manufacturers Association
Date: 5/14/1987
Response: With respect to the status of "Closed-Unacceptable Action" assigned to Safety Recommendation R-83-92, the Safety Board notes your statement that, "this decision was apparently made on the basis that alternative methods of improving tank car/commodity identifications were not credible or easily implemented solutions." The Safety Board points out that, as the recommendation is worded, we recognized that some means other than color coding may be effective in identifying high-risk commodity tank cars. No substantive information was ever received to enable us to come to a determination that the alternative methods were or were not credible or were or were not easily implemented. The decision to classify Safety Recommendation R-83-92 as "Closed-Unacceptable Action" was not based on the credibility or lack thereof of the alternative methods, but rather the lack of substantive information which would suggest to the Safety Board that this project was being actively pursued. As we indicated in our February 18, 1987, letter, the Livingston, Louisiana, accident which generated this recommendation occurred more than 4 years ago, and yet adoption and implementation of an effective system does not appear imminent. We had hoped for more timely action in response to our recommendation. We further indicated in our February 18, 1987, letter, however, that if the Chemical Manufacturers Association (CMA) could provide details regarding the project to use a transponder as a means to improve identification of tank cars (an alternative method briefly mentioned in your November 19, 1986m letter) and an outline of its plan to actively pursue an effective means for tank car identification, we would consider reclassifying the recommendation upon receipt of that information. Again, while we note the assurance that both CMA and the Federal Railroad Administration are conducting a real evaluation of the various alternatives, we have yet to receive the necessary supporting documentation and evidence to justify a reclassification of the recommendation. Safety Recommendation R-83-92 remains in a "Closed-Unacceptable Action" status.

From: Chemical Manufacturers Association
To: NTSB
Date: 3/18/1987
Response: NO RESPONSE NECESSARY.

From: NTSB
To: Chemical Manufacturers Association
Date: 2/18/1987
Response: The Safety Board notes that CMA's position regarding the use of color coding for identifying high-risk commodity tank cars has not changed. The Board notes further that even though CMA indicates that it is continuing to look at ways to improve the identification of tank cars by other means including the use of a transponder, adoption and implementation of an effective system does not appear to be imminent. The Board points out that its investigation of the Livingston, Louisiana, accident, which generated Safety Recommendation R-83-92, occurred more than 4 years ago. Consequently, since the Board had hoped for more timely action in response to its recommendation, we have placed Safety Recommendation R-83-92 in a "Closed--Unacceptable Action" status. If, however, CMA can provide details regarding the project to use a transponder as a means to improve identification of tank cars and an outline of its plan to actively pursue an effective means for tank car identification, the Board would consider reclassifying the recommendation upon receipt of that information. Further, since the Board remains vitally interested in this important safety issue, we look forward to being contacted in the near future to discuss the matter with CMA's Distribution Committee.

From: Chemical Manufacturers Association
To: NTSB
Date: 11/19/1986
Response: CMA HAS REVIEWED R-83-92 AND WE MUST REAFFIRM THE OBJECTIONS WE RAISED IN OUR LETTER OF DECEMBER 8, 1983, REGARDING COLOR CODING OF TANK CARS. OUR REVIEW OD RECENT INCIDENTS HAS NOT REVEALED THAT COLOR CODING WOULD HAVE IMPROVED THE RESPONSE. STILL, CMA IS CONTINUING TO LOOK AT WAYS TO IMPROVE IDEN TIFICATION OF TANK CARS BY OTHER MEANS. ONE OF THE POS SIBILITIES WE ARE REVIEWING IS PLACEMENT OF A TRANSPONDER ON CERTAIN TANK CARS. THIS DEVICE WOULD EMIT A RADIO FREQUENCY THAT WOULD TRANSMIT INFORMATION ON THE CAR'S CONTENTS. WE HAVE DISCUSSED THIS PROJECT WITH THE FEDERAL RAILROAD AD MINISTRATION AND THEIR CONTRACTOR, AND WE WILL CONTINUE TO EVALUATE THIS TECHNOLOGY. WITH RESPECT TO THE CANADIAN TRANSPORT COMMISSION'S REGULATIONS ON COLOR BANDING, WE HAVE BEEN INFORMED BY CP RAIL, AND INFORMALLY BY THE CANADIAN CHEMICAL PROCEDURES' ASSOCIATION, THAT IMPLEMENTION OF THOSE REGULATIONS IS ALREADY BEING CONSIDERED. IN ANY CASE, THE CANADIAN COLOR BANDING SCHEME WOULD NOT ACCOMPLISH THE OBJECTIVES OF YOUR RECOMMENDATIONS BECAUSE IT DOES NOT ADDRESS THE POTENTIAL HAZARDS OF THE MATERIAL IN THE CAR. FOR EXAMPLE, A CAR CONTAINING CHLORINE WOULD BE BANDED IN THE SAME MANNER AS A CAR CONTAINING CARBON DIOXIDE. FINAL LY, WE WOULD WELCOME YOUR SPEAKING TO THE CMA'S DISTRIBUTION COMMITTEE ONTHIS ISSUE. JOE MAYHEW WILL BE CONTACTING YOU IN THE NEAR FUTURE TO ARRANGE A MEETING.

From: Chemical Manufacturers Association
To: NTSB
Date: 8/20/1986
Response: CMA'S INITIAL RESPONSE OF DECEMBER 8, 1983 ATTACHED. NTSB'S FOLLOWUP OF FEBRUARY 21, 1984 ATTACHED.

From: NTSB
To: Chemical Manufacturers Association
Date: 8/20/1986
Response: CMA responded to the Board's recommendation on December 8, 1983, and indicated that after several meetings CMA had concluded that extended color coding was not the best means of improving tank car identification and listed several possible problems with a color coding system. CMA indicated further that it believed there was greater potential for improving identification of tank car contents by improving one or all of the existing systems than by introducing another system of color coding. The Board, however, pointed out in its February 21, 1984, letter, that the existing systems, as currently implemented, have proven to be inadequate as illustrated in the accident at Livingston, Louisiana. Car numbers and placards that were still in place were often obscured by fire, smoke, and wreckage. The Board urged CMA to defer any decision about the merits of color coding until CMA had completed its review. In its February 21, 1984, letter, the Board indicated that staff would be pleased to participate in CMA meetings to discuss this topic. The Board notes, however, that we have never received a response to our February 21, 1984, letter, nor have we been contacted regarding any meetings. In view of the time that has elapsed since that letter, the Board would appreciate an update of CMA's efforts in this area. Pending a response to the Board's inquiry, Safety Recommendation R-83-92 will remain in an "Open-Acceptable Action" status. Since our letter of February 21, 1984, the Board has become aware that a color coding system has been implemented in the Canadian railroad industry. The Board can foresee possible problems with Canadian tank cars being transported in the United States is a corresponding color coding system or, at least, compatible system is not introduced in the United States. The Board urges CMA to address all facets of this issue, and we again offer our assistance in discussing this important subject.

From: Chemical Manufacturers Association
To: NTSB
Date: 12/8/1983
Response: CHEMICAL MANUFACTURERS ASSOCIATION LETTER: AFTER SEVERAL MEETINGS, WE HAVE CONCLUDED THAT EXTENDING COLOR CODING IS NOT THE BEST MEANS OF IMPROVING TANK CAR IDENTIFICATION. THERE ARE AT LEAST 5 SUBSTANTIAL PROBLEMS IN THE COLOR BANDING SYSTEM. TWO OTHER COMPLETE SYSTEMS AND ONE PARTIAL SYSTEM ARE CURRENTLY BEING USED TO IDENTIFY THE CONTENTS FOR TANK CARS. THE PRIMARY SYSTEM INVOLVES THE USE OF PRODUCT-SPECIFIC IDENTIFICATION NUMBER BASED ON THE UN FOUR-DIGIT NUMBER SYSTEM. THIS NUMBER IS ON A TANK CAR, USUALLY IN THE FORM OF A PLACARD, AND IS ON THE SHIPPING PAPERS. THIS NUMBERING SYSTEM ENABLES THE RESPONDER TO IDENTIFY THE SPECIFIC CONTENTS OF EACH TANK CAR AND TO OBTAIN HANDLING AND CONTAINMENT ADVICE. A SECOND AND COMPLEMENTARY SYSTEM IS BASED ON THE TANK CAR NUMBER. THIS NUMBER IS PERMANENTLY APPLIED IN NINE-INCH HIGH LETTERS ON THE CAR SIDES AND FOUR-INCH HIGH LETTERS ON THE CAR ENDS. THE TANK CAR NUMBER WILL LEAD TO THE PROPER SHIPPING PAPER, WHICH CONTAINS PRODUCT, SHIPPER/CONSIGNEE, AND ID NUMBER INFORMATION. THE THIRD SYSTEM IS USED ON ABOUT FORTY OF THE GENERALLY HIGHER HAZARD CHEMICALS. IT INVOLVES STENCILING THE COMMODITY NAMES NEAR EACH END OF THE TANK CAR IN FOUR-INCH HIGH LETTERS. WE ARE CONFIDENT THAT THERE IS MUCH GREATER POTENTIAL FOR IMPROVING IDENTIFICATION OF TANK CAR CONTENTS, BY IMPROVING ONE OR ALL OF THE EXISTING SYSTEMS THAN BY INTRODUCING A FOURTH SYSTEM OF COLOR-BANDING.