Failure Report- Amoco Oil Company 8/18/05

Page 1Official PDFDOT US Department of Transportation PHMSA Pipeline and Hazardous Materials Safety Administration OPS Office of Pipeline Safety Central Region Principal Investigator Karen Butler/James Bunn Region Director David Barrett Date of Report 6/20/2012 Subject Failure Investigation Report – Amoco Insulating Flange Leak, Granger, IN Operator, Location, & Consequences Date of Failure August 18, 2005 Commodity Released Refined Products, Gasoline City/County & State Granger/St. Joseph County, Indiana OpID & Operator Name 395 Amoco Oil Co Unit # & Unit Name 4573 MERRILLVILLE (AMOCO) SMART Activity # 116167 Milepost / Location MP 74.46 Type of Failure Leak Due to Equipment Failure, Insulating Flange Fatalities 0 Injuries 0 Description of area impacted HCA, High Population, Residential Area Total Costs $49,000#

Page 2Failure Investigation Report – Amoco Insulating Flange Leak, Granger, IN Failure Date 8/18/2005 Executive Summary BP Amoco had determined that an insulating flange on the Whiting to River Rouge pipeline at MP 74.46 should be removed. In anticipation of this activity and as part of a newly developed ongoing flange monitoring program, BP Amoco had Delta Environmental complete the installation of 3 monitoring wells near this insulating flange. On August 18, 2005 at 15:00 EDT, a potential leak was detected at one of these three monitoring wells in Granger, IN near MP 74.46. A layer of free hydrocarbon product approximately 0.25 inch thick was detected. The area around the 12 inch insulating flange was excavated and while some old product was found in the soil near the flange, no liquid was initially observed to be leaking from the insulating flange. Upon excavation and after visual inspection, several loose bolts were re-torqued (1/4 to 1/2 turn). On the following morning (August 19, 2005), the insulating flange was observed to be leaking by BP Amoco employees and the pipeline was officially taken out of service. All of the studs (bolts) and nuts were replaced around the flange in an effort to mitigate any possible leak that could have been present. Approximately 21 gallons of gasoline were removed from the ground at MP 74.46. This accident did not result in any injuries or fatalities. The release was discovered in a High Consequence Area. There was no fire or explosion and there were no service interruptions, but supply was of concern at the time of the hydrocarbon detection. The total cost of the accident reported by BP Amoco was $49,000. Contaminated soil was remediated and samples taken. An environmental contractor sampled water from several residential water wells; there was no evidence of hydrocarbon contamination. On August 25, 2005, when supply concerns diminished, the insulating flange was cut out of the pipeline with the newly installed bolts and nuts intact. A straight section of pipe was installed to replace the flange. The insulating flange with the original gasket material (fiber ring) still in place was sent to a metallurgical laboratory for analysis. The pipeline was then backfilled and placed back in service. Samples from the release site were analyzed and the product was determined to be gasoline with high sulfur content. A product similar to that analyzed had last been moved through the pipeline in October of 2003. This information identified the product discovered as being from a historical release. The metallurgical analysis revealed that the fiber ring joint gasket had some evidence of a prior leak. The flange faces had evidence of a possible leak at two different locations. The analysis determined that this leak would have been intermittent, of minimal volume and possibly characteristic of a fugitive emission release. This pipeline system had four previous flange leaks, has sixteen significant water body crossings and transverses highly populated areas, was reported to have been originally constructed in 1953 with 26 buried flange valves and 17 buried insulating flanges, has not been hydrotested entirely, and at the time of this report contains only one remaining buried flange that is scheduled for removal or abandonment in 2013 due to a pipeline replacement project. System Details The Whiting to River Rouge Pipeline is 243 miles long and transports refined petroleum products from the refinery at Whiting, IN to the terminal at River Rouge, MI. The pipeline elevations range from 570 feet to 1,100 feet. The 120 mile portion of the line between Whiting and Colon junction is 12 inch in diameter and the 123 mile portion of the line between Colon Junction and River Rouge is 10 inch in diameter. The pipeline transverses heavily populated areas near Chicago and Detroit with farmland in between. Sixteen significant water bodies are crossed (rivers and canals mainly) on this pipeline. Page 2 of 7#

Page 3Failure Investigation Report – Amoco Insulating Flange Leak, Granger, IN Failure Date 8/18/2005 The pipeline was reported to have been originally constructed in 1953, has not been hydrotested in its entirety, and initially contained 26 buried flange valves and 17 buried insulating flanges. This insulating flange release occurred in the 12 inch diameter portion of the line which BP Amoco reported to contain 17 flanged-end valves (8 above ground and 9 below ground) and 17 underground insulating flange sets. The wall thickness of the pipeline at the insulating flange location was 0.285 inch, but wall thickness ranges in the 12 inch pipeline between 0.281 inch and 0.688 inch. The 10 inch wall thickness ranges from 0.279 inch to 0.500 inch. The 12 inch is comprised of API 5L, grade X42, seamless line pipe that was manufactured by National Tube. The pipeline was joined by the shielded metal arc welding (SMAW) process and was coated with coal tar enamel. The MOP of the line was 1,440 psig and typical operating pressures were reported by BP for the area of the failure to be 270-300 psig. The line was inspected with a high-resolution in-line magnetic flux metal loss tool in 2002 and reassessed in March of 2007. Events Leading up to the Failure BP Amoco informed PHMSA of a buried flange monitoring program on the Whiting to River Rouge pipeline in March of 2005 and provided basic information about the program to PHMSA at a meeting in April of 2005. The flange monitoring program was the result of several historical flange leaks (1987 at Constantine, MI; 1995 at MP 1.09 amounting to 20 barrels; 1997 at MP 4.56 of 6 barrels; 2000 in Chesterton, IN amounting to 0.5 barrels). The flange monitoring program was to include monitoring wells, soil checks for hydrocarbons using a Photo Ionization Detector (PID), immediate excavation of any location that was suspect regarding an active leak, and continued monitoring on a 6 month basis. The operator was in the process of constructing monitoring wells and removing flanges at several locations in fall of 2005 as part of this monitoring program. The program involved flanges associated with 26 buried valves and 17 buried insulating flanges. PHMSA had identified this program initially as insufficient and requested revisions. BP Amoco had already implemented several monitoring wells along the pipeline in August of 2005 prior to the leak. Three monitoring wells had already detected hydrocarbons initially. The first site was excavated, but no hydrocarbons were found. The second site was evaluated and contained evidence of a non-reportable historical leak. The third site was at the buried insulating flange at MP 74.46, and BP Amoco determined that this buried insulating flange should be removed. In anticipation of this activity and as part of an ongoing flange monitoring program, BP Amoco had Delta Environmental complete installation of 3 monitoring wells near this insulating flange. The two north monitoring wells were determined to be free of product. Emergency Response On August 18, 2005 at 3:00 p.m. EDT, a potential leak was detected at one of three monitoring wells in Granger, IN near MP 74.46. The monitoring well was located on the south side of the buried, ring joint, insulating flange. A layer of free hydrocarbon product approximately 0.25 inch thick was found at this third monitoring well. The area around the 12 inch insulating flange was excavated. Product was thought to be present in the soil near the flange, but nothing was visually observed to be leaking from the insulating flange. BP Amoco made all of the necessary agency notifications including a call to PHMSA. BP indicated upon their initial phone call with PHMSA that this was a release of historical nature. The NRC was notified at 5:48 p.m. EDT on August 18, 2005. BP Amoco offered bottled water to residents in the area as originally it was thought that many of the wells in the area were used as drinking water sources for local residents. Delta Environmental Consultants, Inc. (Delta) was commissioned by BP to determine the extent of any soil and groundwater Page 3 of 7#

Page 4Failure Investigation Report – Amoco Insulating Flange Leak, Granger, IN Failure Date 8/18/2005 contamination. Delta collected samples from seven residential water wells that were located within 500 feet of the flanged connection. It was later determined that water from these wells was actually non- potable water and was used for ponds and sprinkler systems. The results of the tests indicated that there was no contamination in the water from the wells. The contaminated soil was removed and placed in a dump truck for disposal (approximately one dump truck load believed to be 10-27 cubic yards), and samples were taken. Summary of Return-to-Service BP Amoco first tightened several bolts on the existing insulating flange even though it did not appear to be leaking at the time of excavation. Since no leaks were observed when the flanged connection was excavated, BP Amoco and their environmental contractor believed that this was high test gasoline from an old leak. On the morning of August 19, 2005, the flange was observed to be leaking by BP Amoco employees and the pipeline was officially taken out of service. All of the studs and nuts were replaced around the flange in an effort to mitigate any possible leak that would have been present. An additional monitoring well had been put in place by this time and no product was found at this new location. BP Amoco re-started flow in the line on August 19, 2005 due to reported supply concerns in the Detroit area. On August 25, 2005, the insulating flange at MP 74.46 was cut out of the pipeline with the original gasket material and newly installed bolts and nuts intact. A straight section of pipe was installed to replace the flange. The insulating flange accompanied by original bolts and nuts and gasket configuration were sent to a metallurgical laboratory for analysis. The line was then backfilled and placed back in service. Investigation Details Immediately after the gasoline was detected at MP 74.46, BP Amoco checked all other underground flange locations for the presence of hydrocarbons. No additional leaks were found. Approximately 21 gallons of gasoline was removed from the ground at MP 74.46. This accident did not result in any injuries or fatalities. The release was discovered in a High Consequence Area. There was no fire or explosion and there were no service interruptions. Supply was of concern at the time of the hydrocarbon detection. BP Amoco completed a short form (due to spill volume) 7000-1 on September 12 of 2005. This was within the required 30 days. The total cost of the accident as reported on this form was $49,000. PHMSA experience indicates that the use of fiber gaskets in the ring joint type flanges is atypical. The use of fiber gaskets versus metal ring gaskets was not explored during the analysis to eliminate any contribution this may have had to leakage. This type of gasket choice may have contributed to other flange connection leaks that occurred prior to 2005. This investigation concentrated in three areas: 1) The development of an enhanced Buried Flange Monitoring Program; 2) Liquid Sample Analysis to determine the age of the release; and 3) Metallurgical Analysis. Buried Flange Monitoring Program The buried flange monitoring program was further refined with PHMSA input and was determined to Page 4 of 7#

Page 5Failure Investigation Report – Amoco Insulating Flange Leak, Granger, IN Failure Date 8/18/2005 have several Phases. Phase I would be similar to what had originally been submitted by BP Amoco with the additional clarification that monitoring wells would be installed on each side of each valve or insulating flange that was stand-alone. PHMSA also requested that upon immediate excavation of any site suspected of active leakage, repair and/or remediation would occur. Phase II of this program was then developed and planned to be implemented over a 2-3 year period. Phase II would require that all buried mainline and takeoff valves would be excavated and information regarding the physical data would be obtained, recorded and reviewed (type of gasket, insulating sleeves and washer details associated with all bolts or just some, etc.); proper torque would be confirmed on all flange bolts; all bolts, metal washers, or nuts would be replaced with new hardware and then torqued with the appropriate tightening sequence; insulating aspects of the flanges would be addressed as necessary; continued hydrocarbon monitoring would take place and any leaking flanges would be investigated and identified by semi-annual surveillance; all data gathered would be reviewed to determine whether a correlation could be documented regarding flange face type, gasket material, or bolt tightness regarding the likelihood of leaks; and that this information would be used to determine “high risk” valve flanges. This program was further developed over time (total time of 7 years starting in 2005) to include the removal of these flanges or buried valves. BP Amoco monitored all underground flange sites on a monthly basis until December 2005. Quarterly monitoring of the sites was initiated in 2006. As of the time of this report, there is one flanged valve that still remains in the system. This flanged valve is located at a Marathon facility in a tank dike and is scheduled for pipeline replacement in 2013 (either the flanged valve will be removed or abandoned in place with a new section of pipeline installed). If a flange is required to be added to the system today for any reason, it is installed as a raised face flange. Liquid Sample Analysis A sample of liquid from the release site was sent by BP to the Whiting Refinery lab and was also analyzed independently by Torkelson Geochemistry. The Whiting Refinery lab looked into component analysis and Torkelson completed a chromatogram and characterized the degree of degradation. The Whiting Refinery lab looked at the concentrations of 1,967 samples that had been in the River Rouge pipeline from 1997 to August 2005. From these samples and that from the release site, it was determined that the sulfur content was the most significant component to determine potential age of the product released. The sulfur content of the sample from the leak site was determined to be 347 parts per million (ppm). A review of sulfur components from the history of the River Rouge pipeline products determined that nothing in the samples had as high of a sulfur concentration since October of 2003. The chromatogram was analyzed to first determine types of products in the sample and then to estimate the time it may have been in the water. The sample itself had lost the “light end” due to weathering processes (examples are evaporation and biodegradation). The gasoline fraction of the sample was determined to be classified as “moderately degraded”. After factoring in the depth of the water in which it was found, Torkelson determined that the degree of degradation was consistent with a product that had been in the subsurface at least several years. Metallurgical Analysis The metallurgical analysis involved visual inspection of the flange with the gasket in original position, Page 5 of 7#

Page 6Failure Investigation Report – Amoco Insulating Flange Leak, Granger, IN Failure Date 8/18/2005 hydro testing of the flange in original configuration (minus original bolts and nuts), a visual review of both sets of nuts and bolts and insulating flange pieces, and a detailed review of the gasket and flange faces. The metallurgical analysis had the following findings: No clear visible evidence was found of a leak when initially reviewed at the laboratory. No evidence of a hydrocarbon leak was found on the studs or nuts (either the original set or the replaced new set). The pipe ends had caps welded on and the flange assembly was pressure tested with water to 1,000 psig and held for a 60 minute test. No evidence of leakage was found. Visual examination of the internal surface of the pipe and flange did not indicate or show evidence of internal corrosion. No internal obstruction was identified that could have caused the ring joint or the spacer to have interfered with an in-line inspection smart pig. When the flange was disassembled, there was some residual evidence of a leak on the flange faces. areas. If the flange was leaking, it was most likely intermittent, very minimal and would be like a fugitive emission. No oil-like deposits were found on either flange face or either groove of the ring joint. The ring joint fiber ring gasket showed evidence of a possible leak. There was a small deposit at the bottom of the outside surface of the gasket ring (6 o’clock) that had a hydrocarbon smell. Findings and Contributing Factors Four previous leaks from buried flanges were identified. Historical releases may be found through buried flange monitoring programs and monitoring wells placed on either side of a buried flange can be helpful in identifying leakage locations. Fiber gaskets may have contributed to the leak history on this pipeline but this was examined. This pipeline has not been hydrotested throughout all segments. Per 49CFR 195.303, hazardous liquid pipeline operators may elect to follow a risk-based alternative to pressure testing of older pipelines. The liquid sample analysis identified that degradation and sulfur component amounts indicate the leak The staining on the flange faces did not line up which could indicate at least two possible leak Page 6 of 7#

Page 7Failure Investigation Report – Amoco Insulating Flange Leak, Granger, IN Failure Date 8/18/2005 was from October 2003 or before. PHMSA rejected BP Amoco’s initial flange monitoring program but cooperation between parties resulted in the removal of all but one buried flange over 7 years. This was accomplished in two phases. At the time of this report, 1 buried flange remains in the system and should be eliminated from service in 2013. It is located in a tank dike and is anticipated to require pipe replacement. The flange leak was eliminated during the bolt replacement or initially upon the torqueing of several bolts. If a flange is required on the River Rouge system today, a raised faced flange is used. The flange connection with new flange bolts and nuts passed a pressure test at 1,000 psi for 60 minutes without leakage. The flange faces had indications of a leak in two different locations. The ring joint fiber ring gasket showed evidence of a possible leak. There was a small deposit at the bottom of the outside surface of the ring gasket (6 o’clock) that had a hydrocarbon smell. If the flange was leaking, it was most likely intermittent, very minimal and would be similar to a fugitive emission. This investigation identifies the need for operators to include in their risk assessment programs elements associated with buried flange valves and buried flanges, insulated or other. It also establishes that insulating buried flange monitoring programs should be employed when pipeline assets have this as an identified risk element. Risk assessment programs should also include an understanding of the original condition of the asset construction such as fiber gaskets and hydrotest conditions. Appendices A Maps and Photographs B NRC Report C Operator Accident Report D Metallurgical Report Page 7 of 7#

Page 8Appendix A Maps & Photographs Overview Map Whiting to River Rouge Pipeline Page 1 of 4#

Page 9Appendix A Maps & Photographs At Fir Road, Looking West The Leaking Flange Excavated Page 2 of 4#

Page 10Appendix A Maps & Photographs The Leaking Flange Facing West Residential Area Looking West Leaking Flange Is Located Bottom Right Page 3 of 4#

Page 11Appendix A Maps & Photographs Looking East to the Leak Site Failure Location (Northwest) Page 4 of 4#

Page 12Appendix B NRC 769581 Report NATIONAL RESPONSE CENTER 1-800-424-8802 *** For Public Use *** Information released to a third party shall comply with any applicable federal and/or state Freedom of Information and Privacy Laws Incident Report # 769581 INCIDENT DESCRIPTION *Report taken at 17:48 on 18-AUG-05 Incident Type: PIPELINE Incident Cause: EQUIPMENT FAILURE Affected Area: GROUND WATER The incident was discovered on 18-AUG-05 at 15:00 local time. Affected Medium: WATER GROUND WATER ____________________________________________________________________________ SUSPECTED RESPONSIBLE PARTY Organization: BP PIPELINES MERRILLVILLE, IN Type of Organization: PRIVATE ENTERPRISE ____________________________________________________________________________ INCIDENT LOCATION 15050 HUNTING RIDGE RD County: ST. JOSEPH City: GRANGER State: IN ____________________________________________________________________________ RELEASED MATERIAL(S) CHRIS Code: GAS Official Material Name: GASOLINE: AUTOMOTIVE (UNLEADED) Also Known As: Qty Released: 0 UNKNOWN AMOUNT Qty in Water: 0 UNKNOWN AMOUNT ____________________________________________________________________________ DESCRIPTION OF INCIDENT THE CALLER IS REPORTING THE DISCOVERY OF A GASOLINE RELEASE INTO GROUND WATER COMING FROM A LEAK IN AN UNDERGROUND PIPELINE. ____________________________________________________________________________ INCIDENT DETAILS Pipeline Type: OTHER DOT Regulated: UNKNOWN Pipeline Above/Below Ground: BELOW Exposed or Under Water: NO Pipeline Covered: UNKNOWN ---WATER INFORMATION--- Body of Water: GROUND WATER Tributary of: Nearest River Mile Marker: Water Supply Contaminated: NO ____________________________________________________________________________ DAMAGES Fire Involved: NO Fire Extinguished: UNKNOWN INJURIES: NO Hospitalized: Empl/Crew: Passenger: FATALITIES: NO Empl/Crew: Passenger: Occupant: EVACUATIONS: NO Who Evacuated: Radius/Area: Damages: NO Length of Direction of Closure Type Description of Closure Closure Closure Air: N Road: N Major Artery: N Waterway: N http://www.nrc.uscg.mil/reports/rwservlet?standard_web+inc_seq=769581[6/6/2012 2:54:04 PM] Page 1 of 2#

Page 13Appendix B NRC 769581 Report Track: N Passengers Transferred: UNKNOWN Environmental Impact: UNKNOWN Media Interest: NONE Community Impact due to Material: NO ____________________________________________________________________________ REMEDIAL ACTIONS EXCAVATED THE LINE / AN INVESTIGATION IS UNDERWAY Release Secured: UNKNOWN Release Rate: Estimated Release Duration: ____________________________________________________________________________ WEATHER Weather: CLEAR, ºF ____________________________________________________________________________ ADDITIONAL AGENCIES NOTIFIED Federal: State/Local: State/Local On Scene: State Agency Number: ____________________________________________________________________________ NOTIFICATIONS BY NRC CG INVESTIGATIVE SVC CHICAGO (CGIS ROA CHICAGO) 18-AUG-05 17:52 DOT CRISIS MANAGEMENT CENTER (PRIMARY) 18-AUG-05 17:52 U.S. EPA V (PRIMARY) 18-AUG-05 17:54 NATIONAL INFRASTRUCTURE COORD CTR (PRIMARY) 18-AUG-05 17:52 NOAA 1ST CLASS BB RPTS FOR IN (PRIMARY) 18-AUG-05 17:52 RSPA OFFICE OF PIPELINE SAFETY (PRIMARY) 18-AUG-05 17:54 IN DEPT ENV MNGMT ATTN: BEAUCHAMP (PRIMARY) 18-AUG-05 17:52 MICHIGAN DEQ ATTN: THOR STRONG (PRIMARY) 18-AUG-05 17:52 ____________________________________________________________________________ ADDITIONAL INFORMATION NONE ___________________________________________________________________________ *** END INCIDENT REPORT # 769581 *** The National Response Center is strictly an initial report taking agency and does not participate in the investigation or incident response. The NRC receives initial reporting information only and notifies Federal and State On-Scene Coordinators for response. The NRC does not verify nor does it take follow-on incident information. Verification of data and incident response is the sole responsibility of Federal/State On-Scene Coordinators. Data contained within the FOIA Web Database is initial information only. All reports provided via this server are for informational purposes only. Data to be used in legal proceedings must be obtained via written correspondence from the NRC. http://www.nrc.uscg.mil/reports/rwservlet?standard_web+inc_seq=769581[6/6/2012 2:54:04 PM] Page 2 of 2#

Page 14Appendix C Operator Accident Report NOTICE: This report is required by 49 CFR Part 195. Failure to report can result in a civil penalty not to exceed $25,000 for each violation for each day that such violation persists except that the maximum civil penalty shall not exceed $500,000 as provided in 49 USC 60122 Form Approved OMB No. 2137-0047 U.S. Department of Transportaion Research and Special Programs Administration Report Date SEP 12,2005 No. 20050257 -- 3875 (DOT Use Only) ACCIDENT REPORT - HAZARDOUS LIQUID PIPELINE SYSTEMS INSTRUCTIONS Important: Please read the separate instructions for completing this form before you begin. They clarify the information requested and provide specific examples. If you do not have a copy of the instructions you can obtain one from the Office Of Pipeline Safety Web Page at http://ops.dot.gov PART A - GENERAL REPORT INFORMATION check: Original Report Supplemental Report Final Report 395 / 1. a. Operator's OPS 5-digit Identification Number (if know) / 2. b. If Operator does not own the pipeline, enter Owner's OPS 5-digit Identification Number (if know) / / c. Name of Operator AMOCO OIL d. Operator street address 28100 TORCH PARKWAY e. Operator address WARRENVILLE DUPAGE IL 60555 City, County, State and ZIP Code 2. Time and date of the accident / / / / / / / / 1500 08 18 2005 hr. month day year 0 3. Location of accident (If offshore, do not complete a through d See Part C.1) IMPORTANT: IF THE SPILL IS SMALL, THAT IS, THE AMOUNT IS AT LEAST 5 GALLONS BUT IS LESS THAN 5 BARRELS, COMPLETE THIS PAGE ONLY, UNLESS THE SPILL IS TO WATER AS DESCRIBED IN 49 CFR 195.52(A)(4) OR IS OTHERWISE REPORTABLE UNDER 195.50 AS REVISED IN CY 2001. 5. Losses (Estimated) Public/Community Losses reimbursed by operator: Public/private property damage $ Cost of emergency response phase $ Cost of environmental remediation $ Other Costs $ (describe) Operator Losses: 0 49,000 a. Latitude: Longitude: 0 (If not available, see instructions for how to provide specific location) b. GRANGER ST. JOSEPH City and County or Parish c. IN State and Zip Code Value of product lost Value of operator property damage d. Mile post/valve station or Survey Station no. (whichever gives more accurate location) Other Costs (describe) Total Costs: $ $ $ $ 0 0 0 4. Telephone report Yes No 49,000 / / / / / / / / 769581 08 18 2005 NRC Report Number month day year 6. Commodity Spilled (If Yes, complete Parts a through c where applicable) c. Estimated amount of commodity involved : a. Name of commodity spilled GASOLINE Barrels b. Classification of commodity spilled: HVLs/other flammable or toxic fluid which is a gas at ambient conditions Gallons (check only if spill is less than one barrel) CO2 or other non-flammable, non-toxic fluid which is a gas at ambient conditons Amounts : Gasoline, diesel, fuel oil or other petroleum product which is a liquid at ambient conditions Crude oil Spilled : 21 Recovered: CAUSES FOR SMALL SPILLS ONLY (5 gallons to under 5 barrels) : (For large spills [5 barrels or greater] see Part H) Corrosion Materal and/or Weld Failures Natural Forces Other Ouside Force Damage Excavation Damage Equipment Incorrect Operation Other (630) 836-3491 PART B - PREPARER AND AUTHORIZED SIGNATURE LARRY S. ABRAHAM (type or print) Preparer's Name and Title ABRAHALS@BP.COM Area Code and Telephone Number (630) 836-3582 Preparer's E-mail Address Area Code and Facsimile Number Authorized Signature (type or print) Name and Title Date Area Cod and Telephone Number Form RSPA F 7000-1 ( 01-2001 ) Page 1 of 4 Reproduction of this form is prmitted. PHMSA Data Facsimile 1 of 4#

Page 15Appendix C Operator Accident Report PART C - ORIGIN OF THE ACCIDENT (Check all that apply) 1. Additional location information Offshore: Yes No (comple d if offshore) a. Line segment name or ID d. Area Block # b. Accident on Federal land other than Outer Continental Shelf Yes No / or Outer Continental Shelf State / c. Is pipeline interstate? Yes No 2. Location of system involved (check all that apply) a. Type of leak or rupture Operator's Property Pipeline Right of Way High Consequence Area (HCA)? Describe HCA 3. Part of system involved in accident Above Ground Storage Tank Leak: Rupture: Pinhole Connection Failure (complete sec. H5) Puncture, diameter (inches) Circumferential - Separation Longitudinal - Tear/Crack, length (inches) Propagation Length, total, both sides (feet) N/A Other Cavern or other below ground storage facility Pump/meter station; terminal/tank farm piping and equipment, including sumps Other Specify: b. Type of block valve used for isolation of immediate section: Onshore pipeline, including valve sites Offshore pipeline, including platforms if failure occurred on pipeline, complete items a - g: 4. Failure occurred on Pipe Seam Sump Valve Welded Fitting Scraper Trap Joint Metering Facility Bolted Fitting Upstream: Manual Automatic Remote Control Check Valve Downstream: Manual Check Valve c. Length of segment isolated Automatic Remote Control ft d. Distance between valves ft e. Is segment configured for internal inspection tools? Yes No f. Had there been an in-line inspection device run at the point of failure? Yes No Don't Know Not Possible due to physical constraints in the system g. If Yes, type of device run (check all that apply) High Resolution Magnetic Flux tool Year run: Low Resolution Magnetic Flux tool Year run: UT tool Year run: Geometry tool Year run: Caliper tool Year run: Crack tool Year run: Body of Pipe Pump Component Repair Sleeve Girth Weld Other (specify) Year the component that failed was installed: / 5. Maximum operating pressure (MOP) a. Estimated pressure at point and time of accident: PSIG b. MOP at time of accident: PSIG c. Did an overpressurization occur relating to the accident? / Yes No Other tool PART D - MATERIAL SPECIFICATION / in. 1. Nominal pipe size (NPS) / In open ditch 1. Area of accident 2. Wall thickness / / in. Above ground 3. Specification 4. Seam type SMYS / Under water Other 5. valve type 6. Manufactured by inches Hard Spot tool Year run: Year run: PART E - ENVIRONMENT in year / / / Under pavement Underground Inside/under building 2. Depth of cover: PART F - CONSEQUENCES 1. Consequences (check and complete all that apply) a. Fatalities Injuries c. Product ignited Yes No d. Explosion Yes No Number of operator employees: 0 0 e. Evacuation (general public only) / / people Contractor employees working for operator: 0 0 Reason for Evacuation: General public: 0 0 Precautionary by company Totals: 0 0 Evacuation required or initiated by public official b. Was pipeline/segment shutdown due to leak? Yes No f. Elapsed time until area was made safe: If Yes, how long? days hours minutes / / hr. / / min. e. Water Contamination: Amount in water Ocean/Seawater Surface 2. Environmental Impact a. Wildlife Impact: Fish/aquatic Bird Terrestrial b. Soil Contamination Yes No If Yes, estimated number of cubic yards: Yes No Yes No Yes No c. Long term impact assessment performed: Yes No d. Anticipated remediation Yes No If Yes, Check all that apply: Surface water Groundwater Soil Vegetation Wildlife Groundwater Drinking water Yes No (If Yes, provide the following) barrels No No No No Yes Yes Yes Yes (If Yes, check below.) Private well Public water intake Form RSPA F 7000-1 ( 01-2001 ) Page 2 of 4 Reproduction of this form is prmitted. PHMSA Data Facsimile 2 of 4#

Page 16Appendix C Operator Accident Report PART G - LEAK DETECTION INFORMATION Yes No 2. Was the release intially detected by? (check one): CPM/SCADA-based system with leak detection H1 - CORROSION 1. External Corrosion 2. Internal Corrosion (Complete items a - e where applicable.) 1. Computer based leak detection capability in place? Static shut-in test or other pressure or leak test Local operating personnel, procedures or equipment Remote operating personnel, including controllers Air patrol or ground surveillance A third party Other (specify) 3. Estimated leak duration days hours PART H - APPARENT CAUSE Important: There are 25 numbered causes in this Part H. Check the box corresponding to the primary cause of the accident. Check one circle in each of the supplemental categories corresponding to the cause you indicate. Secc the instructions for guidance. a. Pipe Coating Bare Coated b. Visual Examination c. Cause of Corrosion Localized Pitting General Corrosion Other Galvanic Atmospheric Stray Current Microbiological Cathodic Protection Disrupted Stress Corrosion Cracking Selective Seam Corrosion Other d. Was corroded part of pipeline considered to be under cathodic protection prior to discovering accident? No Yes, Year Protection Started: / / e. Was pipe previously damaged in the area of corrosion? No / years / / months Unknown Yes => Estimated time prior to accident: / Earthquake Subsidence Landslide Other Washouts Flotation Mudslide Other Thermal stress Frost heave Frozen components Other H2 - NATURAL FORCES 3. Earth Movement => 4. Lightning 5. Heavy Rains/Floods => Scouring 6. Temperature => 7. High Winds H3 - EXCAVATION DAMAGE 8. 9. Operator Excavation Damage (including their contractors/Not Third Party) Third Party (complete a-f) a. Excavator group General Public Government Excavator other than Operator/subcontractor Road Work b. Type: Pipeline Water Electric Sewer Phone/Cable Landowner-not farming related Farming Raildroad Other liquid or gas transmission pipeline-operator or their contractor Nautical Operations Other Open Trench c. Excavation was: Sub-strata (boring, directional drilling, etc...) Yes d. Excavation was an ongoing activity (Month or longer) No If Yes, Date of last contact / e. Did operator get prior notification of excavation activity? Yes; Date received: / / mo. / / day / / yr. No / Notification received from: Excavator Contractor Landowner One Call System f. Was pipeline marked as result of location request for excavation? i. Temporary markings: Flags Stakes Paint ii. Permanent markings: Yes No iii. Marks were (check one) : Accurate Not Accurate iv. Were marks made within required time? Yes No No Yes (If Yes, check applicable items i - iv) H4 - OTHER OUTSIDE FORCE DAMAGE 10. Fire/Explosion as primary cause of failure => Fire/Explosion cause: 11. Car, truck or other vehicle not relating to excavation activity damaging pipe 12. Rupture of Previously Damaged Pipe 13. Vandalism Man Made Natural Form RSPA F 7000-1 ( 01-2001 ) Page 3 of 4 Reproduction of this form is prmitted. PHMSA Data Facsimile 3 of 4#

Page 17Appendix C Operator Accident Report H5 - MATERIAL AND/OR WELD FAILURES Material 14. Dent Body of Pipe => 15. 16. Component => Joint => Gouge Bend Arc Burn Other Valve Fitting Vessel Extruded Outlet Other Gasket O-Ring Threads Other Weld 17. Butt => Pipe Fabrication Other 18. Fillet => Branch Hot Tap Fitting Repair Sleeve Other 19. Pipe Seam => LF ERW DSAW Seamless Flash Weld HF ERW SAW Spiral Other Complete a-g if you indicate any cause in part H5. a. Type of failure: Construction Defect => Poor Workmanship Procedure not followed Poor Construction Procedures Material Defect b. Was failure due to pipe damage sustained in transportation to the construction or fabrication site? Yes No c. Was part which leaked pressure tested before accident occurred? Yes, complete d-g No d. Date of test: / / yr. / / mo. / / day e. Test medium: Water Inert Gas Other f. Time held at test pressure: / / hr. g. Estimated test pressure at point of accident: PSIG H6 - EQUIPMENT 20. Control valve Malfunction of Control/Relief Equipment => Instrumentation SCADA Communications Block valve Relief valve Power failure Other 21. Threads Stripped Broken Pipe Coupling => Nipples Valve Threads Dresser Couplings Other 22. Seal Failure => Gasket O-Ring Seal/Pump Packing Other H7 - INCORRECT OPERATION 23. Incorrect Operation a. Type Inadequate Procedures Inadequate Safety Practices Failure to Follow Procedures Other b. Number of employees involved who failed a post-accident test: drug test: / / alcohol test / H8 - OTHER 24. Miscellaneous, describe: 25. Unknown Investigation Complete Still Under Investigation (Submit a supplemental report when investigation is complete) / PART I - NARRATIVE DESCRIPTION OF FACTORS CONTRIBUTING TO THE EVENT (Attach additional sheets as necessary) Form RSPA F 7000-1 ( 01-2001 ) Page 4 of 4 Reproduction of this form is prmitted. PHMSA Data Facsimile 4 of 4#

Page 18Appendix D Amoco Oil Company Metallurgical Report This document is on file at PHMSA#