UNITED STATES COAST GUARD

INVESTIGATION INTO THE CIRCUMSTANCES SURROUNDING THE COLLISION BETWEEN THE

M/V BALSA 37

THE TUG SEAFARER AND

T/B OCEAN 255,

AND THE TUG CAPT. FRED

BOUCHARD AND T/B NO. 155

ON AUGUST 10, 1993, WITH NO LOSS OF LIFE

CONTENTS

1. COMMANDANT'S ACTION

Comments on Conclusions

Action on Recommendations

2. MARINE BOARD OF INVESTIGATION REPORT FINDINGS OF FACT

Summary

Vessel Data

Personnel License Data

Narrative Descriptions

Weather Conditions

The Channel

The Voyage of the BALSA 37

The Voyage of the SEAFARER tow

The Voyage of the BOUCHARD tow

Fire Fighting

Life Saving and Rescue

Injuries

Drug Testing

Personnel Histories

Recording of Radio Communications

Marine Board Visits

CONCLUSIONS

RECOMMENDATIONS

Iii

1

1

2

6

6

6

6

9

10

12

14

15

16

16

17

17

17

18

23

1

3

4

7

8

ILLUSTRATIONSFire aboard the T/B OCEAN 255

Tug CAPT FRED BOUCHARD and the T/B NO.155

The M/V BALSA 337

Chart showing the area of the collision

Position of the vessels prior to collision

Commandant Washington, D.C. 20593-0001United States Coast Guard Staff Symbol: G—MMI—1

Phone: ( 202)267-1430

Commandant's Action

on

The Marine Board of Investigation convened to investigate thecircumstances surrounding the collision of the BALSA 37, Lloyd'snumber 8511794; the tug SEAFARER, O.N. 532672 and T/B OCEAN 255,O.N. 534910; and the tug CAPT FRED BOUCHARD, O.N. 644119 and T/B

NO. 155, O.N. 603622 on 10 August 1993 with no loss of life

COMMENTS ON CONCLUSIONS

Conclusion 23: Recording the critical radio transmission made on channel 13 in theminutes leading up to the casualty would have assisted in this investigation.Specifically, had Channel 13 been recorded to correlate transmissions to theprecise time of day, the process of reconstructing events leading up to thecasualty would have been enhanced.

Comment : I concur with this conclusion. In ports where Vessel Traffic Servicesare located, Channel 13 is monitored and recorded. These recordings have provenuseful in reconstructing events leading up to a marine casualty.

Conclusion 29: A Vessel Traffic Service (VTS) employing the latest technologymight have assisted in the prevention of this casualty. It could have providedinformation as to the general vessel traffic situation including when, where andthe type of vessel that would be met during the transit. A VTS could havedetermined that the three vessels would meet at the turn and that the BALSA 37 wasproceeding down the center of the channel as it approached the turn. Radiotransmissions could have been monitored raising the question as to the BALSA 37pilot's awareness of the situation. Finally, a VTS could have initiated a call foremergency response at the instant it became apparent the collision was imminent.

Comment : I concur with this conclusion. A properly equipped VTS could haveintervened to prevent this accident. VTS interacts with marine traffic byproviding accurate and complete information to all mariners. VTS watchstanders inthe vessel traffic center respond to developing situations in the waterway andadvise mariners of potentially dangerous conditions. Watchstanders ensure thatmariners have timely, relevant and accurate information to support theirindependent decision

making. If watchstanders determine that proper action is not beingmade to alleviate a dangerous situation, they are empowered todirect vessel movements. A "watchstanderless" system would not havethe capability to interact with traffic or provide third party,independent assessment of the traffic situation. Independentassessment and monitoring of the waterway are critical elements inavoiding future accidents in Tampa Bay.

ACTION ON RECOMMENDATIONS

Recommendation 1: That Commandant develop and implement a merchantmariner's license and document revocation procedure based upon ahistory of confirmed improper actions and/or violations. A system ofpoints should be assessed for violations of the Code of FederalRegulations, proved findings of charges in suspension and revocationproceedings, and civil and criminal convictions. When theaccumulated points exceed a specified level, the Coast Guard shouldautomatically initiate revocation proceedings. It is furtherrecommended that the State of Florida develop a similar system forpilots under its jurisdiction.

Action : I concur with this recommendation. The Coast Guard will seeklegislative authority to establish the recommended "points system."The Oil Pollution Act of 1990 (OPA 90) gave the Coast Guardadditional authority which also addresses the intent of thisrecommendation. OPA 90 requires renewal of certificates of registryand merchant mariner's documents every five years, thus ensuringthat the Coast Guard has interaction with the holders of credentialson a more frequent basis. OPA 90 also authorizes the Coast Guard toreview the National Driver Register for vehicular offenses(including drug and alcohol involvements), to conduct criminalrecord checks, and to revoke the credentials of holders withcriminal convictions or convictions of offenses of the NationalDriver Register Act of 1982. When these OPA 90 authorities areimplemented in regulation, the Coast Guard-will have significantlymore interaction with mariners, will have access to more backgroundinformation on them, and will be in a significantly better positionto protect life, property, and the marine environment throughoversight of merchant mariners.

Recommendation 2: That Commandant develop a new license anddocument application and renewal procedure which will determineand list all prior suspension and revocation actions, and othercriminal convictions and Driving Under the Influence (DUI)convictions. If additional statutory authority is necessary toperform such checks, it should be sought.

ii

I

Action : I concur with the intent of this recommendation. Although the existingapplication and renewal procedures allow for the consideration of priorsuspension and revocation actions, criminal convictions, and DUI convictions,Commandant (G-MVP) has initiated several projects to improve theeffectiveness of these procedures.

A regulatory initiative is underway to provide for verification ofinformation that mariners provide on their applications regardingDUI convictions by obtaining information from the National DriverRegister. Initiating spot checks of criminal records during licenseand Merchant Mariner Document renewals and upgrades will help verifyinformation regarding criminal convictions and may serve as an addedincentive to mariners to include all relevant information on theirapplications. Finally, implementation of the Merchant MarinerLicensing and Document system will allow for the review of anapplicant's history of administrative actions reported byinvestigating officers and will help to verify the information thatmariners provide on their applications.

Although the license application, Form CG-866, specifically requestsinformation regarding suspensions and criminal convictions,Commandant (G-MVP) will revise and clarify portions of theapplication. The question regarding suspensions will be revised toinclude all types of administrative actions, not just suspensions.The question regarding criminal convictions will be reworded toemphasize that DUI convictions are serious offenses that must beincluded on the application.

Recommendation 3: That the Tampa Bay Pilots Association educatemember pilots on the requirements for post-accident chemicaltesting.

Recommendation 4: That the Tampa Bay Pilots Association contractwith a qualified facility to expedite the collection of samples fordrug testing its members on a 24-hour basis.

Action : I concur with the intent of these recommendations. Existingregulations at 46 CFR 4.06.1(e) require marine employers to ensurethat all individuals engaged on board a vessel are fullyindoctrinated in the requirements of serious marine incidentchemical testing, and 46 CFR 4.06-20(c) requires marine employers toensure specimens for drug and alcohol testing are collected as soonas possible. Commandant (G-MMI) will forward a copy of this reportto the Tampa Bay Pilots Association for their information andreview, and will request a written response to theserecommendations.

Recommendation 5: That the Marine Safety Office, Tampa MarineFire Fighting Contingency Plan be reviewed and revised asnecessary to improve communications during such emergencies.

iii

Action : I concur with this recommendation. Review and revision ofunit contingency plans following a major incident is standardprocedure. The importance of the review and revision process will beemphasized in a revision to the Marine Safety Manual, Volume VI,Chapter 8.

Recommendation 6: That, nationwide, Commandant establish means forrecording Channel 13, or the designated bridge-to-bridge channel, incritical waterways where such recording is not currently being doneby VTS or other system. Such recordings frequently are of great valuein casualty investigations.

Action : I concur with the intent of this recommendation. Recordings ofthe voice radio exchanges between vessels have proven valuable inreconstructing the events leading up to marine casualties. However,this benefit of monitoring and recording Channel 13 must be comparedwith the costs of incorporating the additional recording capabilityinto the National Distress and Safety System. The cost of addingChannels 13 and 67 guard receivers, data links and recording capacityis appreciable. These costs are not readily justified when consideringthat the majority of bridge-to-bridge transmissions are "businesstransactions" between vessels and that shore stations have little needto communicate on bridge to bridge frequencies.

Recommendation 7: That Commandant explore alternative or additionalnoisemaking devices on Personal Flotation Devices whose operationwould be unaffected by oil or chemicals in the water.

Action: I partially concur with this recommendation. Whistles areinexpensive, lightweight, inherently maintenance-free, and normallyeffective. The alternatives to whistles that are currently availableinvolve some sort of power source and are significantly more bulky andexpensive. Therefore, noisemaking devices other than whistles areunlikely to be practical. Whistle manufacturers often bring theirproducts to the attention of the Coast Guard. As they do, we will alertthem to the possible problems of fouling.

Recommendation 8: That Commandant initiate a review of the marinefirefighting capability in the nations ports, on its waterways, andoffshore. Included in a national review should be the consideration ofthe use of portable, high capacity foam equipment.

Monitors, prime movers, and foam stocks could be strategically located inport areas and transported by land, sea, or air to areas within a port, toremote waterway sites, or offshore. Within certain small ports, suchequipment could substitute for fireboats, precluding a major capitolinvestment.

iv

Readily deployable, high capacity, portable firefighting equipment could be thedeciding factor in preventing or minimizing the further release of pollutants tothe environment in any number of casualty scenarios.

Action : I concur with the intent of this recommendation. In its recent study AReassessment of the Marine Salvage Posture of the United States (1994), theMarine Board of the National Research Council (NRC) found that the marinefirefighting capability of the United States has improved significantly since1982. However, the NRC report also found that "... gaps still exist in marinefirefighting experience and capability in certain port areas ...". Commandant(G-MPS) will initiate a study of the firefighting capabilities within thenations ports and harbors, and identify any port areas where the marinefirefighting capability is deficient. Included in this study will be ananalysis of any equipment needed to improve the firefighting capabilities forport areas.

To support local authorities in meeting their responsibilities for marinefirefighting, the Coast Guard participated in a joint USCG/National Fire ProtectionAssociation (NFPA) working committee to develop the publication NFPA 1405 "Guide forLandbased Fire Fighters Who Respond to Marine Vessel Fires 1990". This document notonly addresses specific fire fighting safety concerns and techniques for vesselfires, it also discusses command and control functions which identify the localCaptain of the Port (COPT)/Group Commander as the marine technical advisor to localfire departments responding to vessel fires. In offshore areas, it is theresponsibility of the vessel's owner and/or operator to provide any necessary firefighting capability.

During the development of the Area Contingency Plans (ACP's), COTP's/On SceneCommanders's (OSC's) include the firefighting resources available within theirzone. The policy guidance on ACP development is being revised to include therequirement for an assessment of the marine firefighting capabilities withineach COTP/OSC zone, the identification of any shortfalls in capabilities, andthe development of procedures to deal with any deficiencies.

Also, as part of the Vessel Response Plans and Facility Response Plans reviewprocess, each response plan is evaluated to ensure resources and proceduresare identified to mitigate or prevent any discharge in the event of a vesselor facility fire situation. If this information is not included, the plan isreturned for revision.

FINDINGS OF FACT

SUMMARY

Before dawn on 10 August 1993, the tug CAPT.FRED BOUCHARD was pushing the petroleum-laden barge B. NO. 155, and was inbound inEgmont Channel, Tampa Bay, Florida. An assisttug, the EDNA ST. PHILLIP, was made up to theB. NO. 155, as the CAPT. FRED BOUCHARDhad suffered a starboard engine casualty. TheIntegrated Tug Barge SEAFARER and OCEAN255, with a mixed cargo of petroleum products,was also inbound and following astern of theBOUCHARD tow. The M/V BALSA 37, havingdeparted Port Manatee, Florida was outbound inMullet Key Channel carrying 6000 metric tons ofphosphate product. Under fair skies, the wind wasnortheast at 12 knots and visibility was 10 to 12miles. The tide was flooding at 0.7 knots.

The Egmont Channel and Mullet Key Channelmeet at a turn just east of buoys 19 and 20. Asthe two inbound tows approached Mullet KeyChannel, the SEAFARER tow commencedovertaking the BOUCHARD tow. The master ofthe SEAFARER did not communicate to theapproaching BALSA 37 his intentions toovertake. This overtaking would place all threevessels in close proximity near the turn. As thetwo inbound tows approached buoys 19 and 20,the SEAFARER tow was near the centerline ofthe channel with the BOUCHARD tow astern andto starboard.

The outbound BALSA 37 steered a courseplacing it near the centerline of Mullet KeyChannel. The BALSA 37 and the SEAFARERtow had made a port-to-port meeting agreement.However, as the vessels approached, each camehard to port to avoid collision. At approximately0545 (all times EDT unless otherwise noted), thestarboard bow of the OCEAN 255 struck thestarboard side of the BALSA 37. The impactruptured the No. 1 starboard tank of the OCEAN255, igniting its cargo. The SEAFARER crewabandoned ship and was rescued by the TampaBay pilot boat MANATEE. The BALSA 37suffered extensive damage along its starboardside and cargo holds. The BALSA 37 separatedfrom the OCEAN 255 and then collided with theBOUCHARD tow less than one minute later,rupturing the B. NO. 155's No. 1 port tank. 5000barrels of No. 6 oil were released. The collisiondamaged the bow of the BALSA 37, rupturing itsforepeak tank.

The SEAFARER tow grounded near Fort DeSoto Park. Its No. 6 starboard tank exploded andthe fire burned out of control until extinguished bythe Tampa Fire Department that evening. TheBALSA 37 was grounded south of the channelnear Egmont Key to prevent sinking. TheBOUCHARD tow proceeded to anchorage nearthe Sunshine Skyway Bridge while continuing torelease oil.

Fire aboard the T/B OCEAN 255

1

DWilliams2

Rectangle

PERSONNEL LICENSE DATA

Thomas A. Baggett: Master of ocean steam ormotor vessels of not more than 500 gross tons;also, first class pilot steam or motor vessels ofany gross tons upon Galveston Bar andchannels to Texas City, Galveston andturning basin, Houston, Texas; LowerMississippi River from Baton Rouge,Louisiana Railroad and Highway Bridge (mile234 AHP) to sea via South Pass; via mainship channels to Tampa, Port Tampa, WeedonIsland and Port Manatee, Florida; Also, radarobserver (unlimited) expires January 1997.Issue number 8-9. Issue date 9 April 1993,Miami, Florida.

: Repubilic of thePhillipines Master Mariner License onmerchant marine vessels of any gross tonsupon the waters of any sea or ocean,Certificate number 3793 issued 26 May1980

Charles C. Chapman: Master of near coastalsteam or motor vessels of not more than 1600gross tons; also, master of near coastal steamor motor vessels of less than 200 gross tons(restricted to uninspected towing vessels); also,operator of uninspected towing vessels uponthe Great Lakes and inland waters; also, firstclass pilot of tug/barge combinations of anygross tons, upon the waters of PortEverglades main ship channel, Ft. Lauderdale,Florida; also, first class pilot of steam ormotor vessels of any gross tons, upon theMississippi Sound from the sea buoy HornIsland Pass via Pascagoula Channel to thejunction of Bayou Casotte via Bayou Casotteto the turning basin; also, first class pilot ofvessels of any gross tons upon the waters ofTampa Bay main ship channel from Egmontsea buoy to and including Sparkman/Yborchannels, also, from Cut "G" through Cut"K", Old Port Tampa, excluding Rattlesnake;also radar observer (unlimited) expires May1994. Issue number 4-4. Issue date 1 August1989, Miami, Florida.

: Mate oceans steam ormotor vessels of not more than 1600 grosstons; also, first class pilot of vessels of anygross tons upon the waters of Tampa Baymain ship channel from Egmont Sea Buoy toand including Sparkmarl/Ybor channels; also,from Cut "G" through Cut "K", Old PortTampa, excluding Rattlesnake Channel; also,radar observer (unlimited) expires May 1995.

Issue number 2-2. Issue date 15 October1990, Baltimore, Maryland.

Robert W. West: Operator of uninspectedtowing vessels upon near coastal routes.Issue 3-3. Issue date 7 July 1989, Baltimore,Maryland.

John M. Wrasse: Chief mate of steam or motorvessels of not more than 1600 gross tons uponoceans; also, second mate of steam of motorvessels of any gross tons upon oceans; also,chief mate of ocean steam or motor vessels ofany gross tons; also, first class pilot of steamor motor vessels of any gross tons upon thewaters of Tampa Bay and its tributarties;also radar observer (unlimited) expiresNovember 1995. Issue number 1-2. Issued 27February 1987, Miami, Florida. (Note:license had expired at time of casualty).

NARRATIVE DESCRIPTIONS

a. Weather Conditions

The weather prevailing in lower Tampa Bay,Florida at the time of the collisions was clear,predawn darkness without fog, haze or other localenvironmental impairment to visibility. Ananemometer located near the Sunshine SkywayBridge indicated that the wind at the time of theaccident was east northeast at 12 knots. CoastGuard surface and air rescue units made thefollowing observations at the scene within 30minutes of the accident:

CloudCover:Visibility:Precipitation: Wind:Seas:SeawaterTemperature:

Clear 10 nauticalmiles None Eastnortheast at 10 to15 knots .3tolm(lto3ft)

Tidal currents in lower Tampa Bay aremeasured by a system of sensors maintainedby the Greater Tampa Bay Marine AdvisoryCouncil's Physical Oceanographic RealTimeSystem (PORTS). The system includes a tidalcurrent sensor located on the bottom of themain ship channel below the northbound laneof the center span of the Sunshine SkywayBridge. PORTS system data has beenextrapolated to other locations in lowerTampa Bay, based on previous observationstaken at these locations.

A tidal current of 0.8 knots towards 071°T wasprojected for Mullet Key Channel at 0500 on 10August 1993. By 0600, the PORTS dataprojeOted a tidal current of 0.7 knots towards070 T for Mullet Key Channel.

b. The Channel (See Fig. 1)

Egmont Channel's project width is 213 meters(700 feet) with a project depth of 14 meters (45feet). It intersects Mullet Key Channel at an angleof 23 degrees. This junction occurs in theproximity of Mullet Key Channel Lighted Buoy19 and Mullet Key Channel Lighted Bell Buoy20. Mullet Key Channel has a project width of183 meters (600 feet) and depth of 13 meters (43feet).

To facilitate vessel traffic through the bend inthe channel, the federal project provides for awidener on the channel's north side. Buoy 19 islocated on the edge of the widener. This widenerallows outbound vessels to provide additionalsea room for inbound vessels to make the turnfrom Egmont Channel into Mullet Key Channel

The charted positions of Mullet Key ChannelLighted Buoy 19 and Lighted Bell Buoy 20

have a tolerance radius of approximately 46 meters(150 feet). This provides for a nominal distance ofapproximately 305 meters (1,000 feet) betw&enbuoys 19 and 20, on a bearing of 193/013 T.Considering the watch circle of each buoy, thedistance between the buoys could vary fromapproximately 230 meters (750 feet) to 410 feet(1,350 feet). However, those conditions thattypically cause a buoy to move within its watchcircle, wind and current, would act in the samerelative direction, causing buoys 19 and 20 tomove relative to each other and therefore maintainan approximate distance from each other of 300meters (1,000 feet).

On 10 August 1993, following the casualty, theAids to Navigation Team from Coast GuardGroup St. Petersburg conducted position checks ofEgmont Channel Lighted Buoy 17 (Light List#18560), Lighted Buoy 18 (Light List #18565),Mullet Key Channel Lighted Buoy 19 (Light List#18580), Lighted Bell Buoy 20 (Light List#18585), Lighted Buoy 21 (Light List #18600),Lighted Buoy 22 (Light List #18605), LightedBuoy 23 (Light List #18610), Lighted Buoy 24(Light List #18615), Mullet Key Range FrontLight (Light List #18590) and Mullet Key RangeRear Light (Light List #18595). All aids were onstation and watching properly.

c. The Voyage of the BALSA 37

The M/V BALSA 37 is a 105m (345'), 4,337gross ton freightship with direct diesel propulsion.The ship is registered in the Philippines and has aFilipino crew.

The BALSA 37 departed Port Manatee, Floridaat 0445 on 10 August 1993 bound for Cartegena,Columbia laden with 6000 metric tons ofphosphate products in bulk in holds No. 1 and 2.The mean draft was 6.7m (22'3") with no trim.All machinery, navigation equipment andnavigation lights had been tested prior todeparture and were working properly.

Pilot Baggett boarded the vessel at 0440. He helda master-pilot conference with Captain ato discuss the order in which lines would bereleased, since the vessel would be undockedwithout the assistance of a tugboat. Pilot Baggettwas familiar with the handling characteristics ofthe BALSA 37, having shifted the vessel theprevious day. The last line was let go at 0445.Pilot Baggett, Captain , the third mateand helmsman were present on the bridge as theBALSA 37 got underway. The third mate waspositioned at the engine order telegraph.

The BALSA 37 passed under the SunshineSkyway Bridge at 0524, and was proceeding at itssea speed of 11 knots. Captainsecured the anchor watch and ordered the chiefmate to prepare the vessel for sea. Pilot Baggettadvised Captain that a bow look-outwas not needed, because the visibility was good.Captain secured the bow lookout andwent below. At approximately 0535, the BALSA37 was near Buoy 23 and the chief mate relievedthe third mate at the engine order telegraph.

The inbound M/V ATLANTIC STAR passed theBALSA 37 between buoys 21 and 23. PilotBaggett and the pilot of the ATLANTIC STARboth testified that the BALSA 37 was to the northside of Mullet Key Channel at that time, passingwithin 100 feet of Buoy 21.

Passing abeam Buoy 21, the BALSA 37Ocontinued on its previous course of 261 T.However, Pilot Baggett Obelieved he hadordered a change to 262 T. which would bringthe ship closer to Buoy 19 at the turn.

With the BALSA 37 in the vicinity of Buoy 21,Pilot Baggett looked at the radar and observedtwo vessels (the BOUCHARD and SEAFARERtows) with one following astern of the other nearBuoy 18 in Egmont Channel. Pilot Baggett thenvisually observed the red sidelights of thesevessels.

Two minutes after passing Buoy 21, Pilot Baggettreceived a radio call from Pilot Wrasse on theBOUCHARD tow proposing a port-to-portmeeting. Pilot Baggett agreed. Approximately oneminute later, Pilot Baggett was called by CaptainChapman on the SEAFARER, also proposing aport-to port meeting. Pilot Baggett again agreed.

Pilot Baggett apparently had not heard a radioconversation during which Captain Chapmaninformed Captain Wrasse that the SEAFARERtow was overtaking the BOUCHARD tow.

Midway down Mullet Key Channel, Pilot Baggettlooked back at the Mullet Key range lights, whichhe observed to be open slightly to the north. Hethen made a radio call to the pilot boat operatorand arranged for his pickup from the BALSA 37and transfer to his next assignment. Pilot Baggettremained unaware that the SEAFARER tow wasclose to the center of Egmont Channel, and wasovertaking the BOUCHARD tow.

Two minutes later, Pilot Baggett made a radiocall asking whether one of the approachingvessels had an assist tug. Captain Baggettheard no reply and did not call again.

At approximately 0545, Pilot Baggett observed agreen sidelight crossing his bow. The bows of theBALSA 37 and the SEAFARER tow were about305 meters to 425 meters (1000 to 1400) feetapart and closing at a relative rate ofapproximately 19 knots. Pilot Baggettimmediately called on Channel 13 to determinethe intentions of the SEAFARER tow. CaptainChapman replied that they had agreed to a port-to-port meeting. Pilot Baggett urgently repliedthat it needed to be starboard-to-starboard.Captain Chapman radioed he would attempt toturn to port.

Pilot Baggett first ordered the BALSA 37rudder left 10 degrees, followed by 20 degrees.He directed the chief mate to call

the captain, and ordered the rudder hard to portand the engines full astern. The helmsmanexecuted the helm commands but the chief matedid not pass the engine orders because he hadexited the bridge to alert Captain who was below.

At approximately 0545, the starboard bow of theOCEAN 255 struck the starboard side of theBALSA 37 at the break of the forecastle. Theforce of the impact caused the BALSA 37 to sheerto port and then back to starboard. Sparks andflames erupted and continued as the Ocean 255raked the starboard side of the BALSA 37. Blacksmoke filled the pilothouse of the BALSA 37, andthe smoke detection system automatically activatedthe general alarm. After the vessels separated, theSEAFARER tow continued past the starboardquarter of the BALSA 37. The collision causedextensive damage to the BALSA 37's hull andsuperstructure.

As the SEAFARER tow moved away, PilotBaggett observed that the BALSA 37 was headedtoward collision with the BOUCHARD tow. In anattempt to make a port-to-port passing, Pilot

ordered the helm hard to starboard, butthere was too little time to regain control.

Less than one minute after striking the OCEAN255, the bow of the BALSA 37 impacted the portbow of the B. NO. 155, creating a large notch inway of the No. 1 port cargo tank. The BALSA 37then came out of the notch and fell astern of theBOUCHARD tow. Captain arrived onthe bridge and ordered all hands to muster for theemergency and had Pilot Baggett take the ship toanchorage near Egmont Key. He then confirmedwith Pilot Baggett that a distress call had been sent.

As the BALSA 37 proceeded toward anchorage,the chief mate opened the cargo hatches andobserved sea water in both cargo holds. Soundingsalso revealed water in the forepeak tank. Captain

, concerned for the vessel's stability,grounded the BALSA 37 near Egmont Key.

The ship remained aground for several days untillightering and salvage operations could becompleted. It was later moved to Port Manateefor damage survey and temporary repairs.

10

d. The Voyage of the SEAFARER Tow

The SEAFARER tow was an Integrated TugBarge of the "dual mode" type, operating in the"push" mode (see Navigation and VesselInspection Circular 2-81 for definitions). Itconsisted of the SEAFARER, a 37.2m (122') tugwith diesel reduction propulsion, and the tankbarge OCEAN 255 with a length of 166m (546')and gross tonnage of 14,679. The tug-bargeconnection was an "Intercon System" (ram andladder, electric-hydraulic) which provides a singledegree of freedom allowing the tug to pitch aboutthe transverse connection between the tug andbarge. The OCEAN 255 had a sailing draft of9.4m (31') forward and 9.1m (30' aft). The lengthoverall of the integrated tow was 195.4m (6412).

The SEAFARER tow departed Pascagoula,Mississippi at 1524 CDT on 8 August 1993carrying 236,000 barrels of petroleum products,including gasoline and Jet-A fuel, bound forGATX Terminal, Tampa, Florida. Mate

assumed the watch at 0145 the morningof 10 August. The SEAFARER tow wasapproximately 23 miles from the Tampa Bay seabuoy, approaching from west northwest at a speedof 9.7 knots. The weather was fair withnortheasterly winds at 10 to 15 knots, seas in theGulf northerly at two to three feet, and visibility of10 miles. At 0300, tests of the SEAFARER'ssteering, whistle and radios were completed andlogged. All propulsion machinery and navigationlights were functioning properly.

At 0406, the SEAFARER tow passed the sea buoyand Mate called Tampa Bay VesselTraffic Advisory to report his sea buoy time,Skyway Bridge ETA of 0610 and GATX TerminalETA of 0930. The Vessel Traffic Advisorynotified Mate that the tanker M/VSPRAY would be outbound during the sameperiod. Mate then observed theBOUCHARD tow which was entering the channelinbound from the north. He also overheard radioconversations wherein the master of theBOUCHARD tow stated that his starboard mainengine was shut down and that he intended to takean assist tug alongside. The SEAFARER tow metthe SPRAY in a portto-port meeting near buoys 5and 6. At that time, the BOUCHARD tow wasslightly less than two miles ahead, proceeding at aspeed of approximately six knots. Matealso determined from radio conversations

ll

:: $

:Rf

that the BALSA 37 was outbound with PilotBaggett aboard.

At approximately 0500, the SEAFARER towpassed buoys 9 and 10, with the BOUCHARD towabout 1.5 miles ahead. ATLANTIC STAR wasapproaching the SEAFARER tow from astern at aspeed of approximately 15 knots. Mate agreed to have the ATLANTIC STAR overtake theSEAFARER tow. When the SEAFARER tow wasapproximately one mile astern of the BOUCHARDtow, Mate slowed. Pilot Wrasse calledMate and asked if he intended toovertake. Mate declined to overtakebecause he did not want to pull back in front of thefaster ATLANTIC STAR.

Captain arrived in the lower pilothouseto relieve Mate of the watch as theSEAFARER tow was being overtaken by theATLANTIC STAR approximately three-quartersof a mile from buoys 15 and 16.

Mate briefed Captain Chapman on thetraffic situation. In a radio conversation with PilotWrasse, Mate declined a second offerto overtake the BOUCHARD tow. Meanwhile, thedistance between the BOUCHARD tow and theSEAFARER tow was closing to less than one-halfmile.

At approximately 0539, Captain Chapmanassumed the watch. Captain Chapman told Mate

to go below for breakfast and return, asit was his policy to have a second person withpilotage endorsements on the bridge when inpilotage waters. At Captain command,Mate advanced the throttles to fullahead as he left the pilothouse. Captain Chapmanwas at the helm (the "helm" is a hand-held deviceon a cable which allows the operator to moveabout the pilothouse) and was referring to buoys tonavigate within the confines of the channel.Captain Chapman steered the SEAFARER towtoward the center of the channel to pass theBOUCHARD tow. He used small ruddermovements to line his tow up with the port side ofthe B. NO. 155. The vessel's radar was operatingin the "heads up" mode, 3 mile range, and wasfunctioning properly. He estimated theSEAFARER's speed at eight knots and the speedof the BOUCHARD tow at six knots.

11

As the SEAFARER tow began to draw closer tothe BOUCHARD tow, Captain Chapmanoverheard Pilot Wrasse call Pilot Baggett on theBALSA 37 and arrange a port-to-port meeting.With the bow of the OCEAN 255 off the portquarter of the BOUCHARD tow, Captain

called Pilot and requestedpermission to overtake. Pilot consented tothe overtaking as long as it was agreeable withPilot . Without unkeying the radio, Captain

called Pilo who replied that aport-to-port meeting was agreeable. Captain

did not communicate to Pilot that the SEAFARER tow was in the process ofovertaking the BOUCHARD tow.

While the SEAFARER tow was passing theBOUCHARD tow, Captain Chapman estimatedthat the starboard side of the B. NO. 155 was15.2m (50 feet) from the starboard side of thechannel. He estimated the starboard side of theOCEAN 255 was a barge width (85 feet) fromthe port side of the B. NO. 155. CaptainChapman did not detect or consciously correctfor any hydrodynamic interaction between thetwo tows.

When the bows of the two barges were nearlyabeam, the tows were approximately six-tenths ofa mile west of buoys 19 and 20, and the BALSA37 was approximately eight-tenths of a mile eastof buoys 19 and 20. At this point, CaptainChapman heard Pilot Baggett call and inquirewhether the SEAFARER tow had an assist tug.Captain Chapman replied he did not, but that theBOUCHARD tow did.

Mate was below decks forapproximatelyfourminutes. Whenhe returned, afew minutes prior to the collision, he could see thebow of the B. NO. 155 to starboard and slightlyastern. He observed that the two tows were notparallel, with the B. NO. 155 oriented slightlytowards the starboard (south) side of the channel.Captain Chapman was watching the approachingBALSA 37. He intended to fall back in front ofthe BOUCHARD tow as he completed his turninto Mullet Key Channel. Both Captain Chapmanand Mate felt that the SEAFARER towwas steering a steady course.

Mate checked the radar andvisually observed the approaching

BALSA 37. He felt that the BOUCHARD tow hadcleared astern. He and Captain Chapman observed thatthe BALSA 37 was close to the point where it shouldstart its turn into Egmont Channel. He asked CaptainChapman whether the approaching vessel was PilotBaggett's ship. Captain Chapman replied that it was andthey had agreed on a port-to-port meeting. Both menwatched the BALSA 37, anticipating its turn at anytime. Captain Chapman was expecting the BALSA 37to make a hard turn into Egmont Channel. Mate

once again checked the radar. It wasapparent that the BALSA 37 had passed the point whereit had to make its turn. While Captain Chapman wasreaching for the radio, Pilot Baggett urgently called asto SEAFARER's intentions. Captain Chapman repliedthat they had agreed on a port-to-port meeting. By now,the BALSA 37 and the SEAFARER were beyond thepoint of "in extremis." Pilot Baggett replied that theywould have to make it a starboard-to-starboard meeting.Captain Chapman replied that he would come to port.

Captain Chapman pulled the throttles to full asternand gave hard left rudder. He sounded the generalalarm, however, the danger signal was notsounded. Seconds later, the SEAFARER towcollided with the BALSA 37.

As the bow of the OCEAN 255 passed alongthe side of the BALSA 37, tremendous heat andsparks were generated and fire erupted, fueled bycargo spilling from the OCEAN 255's rupturedNo. 1 starboard tank. After separating from theBALSA 37, a wall of fire rolled aft along the deckof the OCEAN 255 toward the SEAFARER.The BALSA 37 passed off the starboard side andaft of the SEAFARER.

Captain Chapman attempted to get theSEAFARER out of the notch of the OCEAN255. Using the wheelhouse controls, he wasunsuccessful in withdrawing the rams whichconnect the tug and barge.

Captain Chapman could feel heat and hearcrackling sounds from the flames approachingthe pilothouse. Heavy black smokeencompassed the outside of the pilothouse. Inan attempt to place the SEAFARER towaground on the sandy bottom adjacent MulletKey, Captain Chapman placed the throttlesahead slow. The rudder position remained hardleft.

When the general alarm sounded, the crew of theSEAFARER gathered in the galley and donnedtheir life jackets. They also carried survival suits.They were not able to gather at their designatedmuster station on the boat deck because ofsmoke.

Captain Chapman and Mate left thewheelhouse, checked the upper decks, then foundall personnel accounted for and waiting in thegalley. Captain Chapman and Mate decided to try one more time to get out of thenotch. As they reached the last flight of stairsleading to the lower wheelhouse, CaptainChapman could see the wheelhouse lit up fromthe fire and could hear windows exploding andglass hitting the deck. The two men abandonedtheir attempt to enter the pilothouse and returnedto the galley.

By now, the smoke had increased to the point thatpersonnel at deck level were having difficultyseeing and breathing. The liferaft wasinaccessible because of its location on the upperdeck. Captain Chapman ordered the crew toabandon ship and they proceeded aft on the portside and jumped overboard. Captain Chapmanwas unsure whether the OCEAN 255 had yetgrounded.

When the crew entered the water, they found thatthere was less smoke near the surface andbreathing was easier. They could see smoke andflames above them. Captain Chapman heard arumbling sound from the OCEAN 255 and twoor three explosions. Approximately 12 to 14minutes after the collision, Captain Chapmanheard one loud explosion which was preceded bya loud "hissing" sound (this was assumed to bethe explosion which blew the top off of the No. 6starboard cargo tank).

The crew was later rescued by the pilot boatMANATEE. The OCEAN 255 groundedapproximately 600 yards south southwest of theFort De Soto pier on Mullet Key. The fireburned throughout the day until extinguished latethat night by the Tampa Fire Department. TheSEAFARER suffered extensive fire damage,and the OCEAN 255 was a constructive totalloss.

e. The Voyage of the BOUCHARD Tow

The BOUCHARD tow consisted of the CAPT.FRED BOUCHARD, a 36.6m (120') tug withdiesel reduction propulsion, and the

tank barge NO. 155 with a length of 135.9m(446') and gross tonnage of 9,262. The tug wasmade up with wires in the notch. TheBOUCHARD tow had a length overall of 167.6m(550'). The B. NO. 155 had a sailing draft of8.18m (26'6") forward and 8.18m (26'08") aft.

The BOUCHARD tow departed Good Hope,Louisiana at 0700 CDT, 7 August 1993, boundfor Berth 9, Port Manatee, Florida, with a cargoof 120,000 barrels of No. 6 oil.

At approximately 1615 on 9 August, theengineer on watch shut down the starboardmain engine due to a loss of crankcase vacuum.

At 0315 on 10 August, the BOUCHARD tow wasproceeding toward the entrance of Tampa Bay. Asa result of the engine casualty, Captain Westarranged for the assistance of the tug EDNA ST.PHILLIP during the transit to Port Manatee. TheEDNA ST. PHILLIP made up to the starboardbow of the B. NO. 155.

At 0500, Pilot Wrasse boarded the BOUCHARDtow in the vicinity of Buoy 12 for the transit intoTampa Bay. Captain West remained at the helmthroughout the transit while Pilot Wrasse managedthe radios and advised Captain West with regardto navigation.

After hearing the pilot on the ATLANTIC STARmake arrangements to overtake the SEAFARERtow, Pilot Wrasse also made an arrangement to beovertaken by the ATLANTIC STAR.

Pilot Wrasse then called the SEAFARER tow andstated that the BOUCHARD tow was only makingsix knots, and the SEAFARER tow could overtakeif desired. Mate replied they weren'tgoing much faster, approximately eight knots, andconsidering the heavy outbound vessel trafficahead, they would not overtake.

The M/V ATLANTIC STAR overtook theBOUCHARD tow west of buoys 17 and 18.Pilot Wrasse once again called the SEAFARERtow and offered them the opportunity to overtake.Mate replied they would not at thattime, but would let the BOUCHARD tow knowwhen they would be overtaking.

13

Just after the BOUCHARD tow had passed abeamEgmont Key Lighthouse, Pilot Wrasse called PilotBaggett on the BALSA 37 and arranged a port-to-port meeting. At approximately 0540, with the bowof the B. NO. 155 approximately eight-tenths of amile from buoys 19 and 20, Captain West felt theinteraction of the OCEAN 255, which was startingto overtake. Captain Chapman then calledindicating his position off the BOUCHARD tow'sport quarter and his desire to overtake. PilotWrasse consented to the overtaking as long as itwas agreeable with Pilot Baggett. Pilot Wrasseheard Captain Chapman call Pilot Baggett on theBALSA 37 immediately and arrange for a port-to-port meeting, but he heard no discussion of theovertaking in progress. Neither Captain West norPilot Wrasse initiated any subsequentcommunication with Pilot Baggett to alert him thatthe SEAFARER tow was in the process ofovertaking the BOUCHARD tow.

Captain West and Pilot Wrasse estimated that thestarboard side of the B. NO. 155 wasapproximately 23 to 30 meters (75 to 100 feet)from the channel boundary during the overtaking.They estimated that the OCEAN 255 was about abarge width (22.6 [74 feet]) from the B. NO. 155.

As the stern of the OCEAN 255 came abeam ofthe bow of the B. NO. 155, Captain West detectedand corrected for a hydrodynamic interaction thatcaused the B. NO. 155 to sheer toward Buoy 20.Pilot Wrasse observed that the distance betweenthe barges was increasing.

At approximately 0544, with the bow of the B.NO. 155 approximately 400 meters (1,300 feet)from Buoy 20, Pilot Wrasse and Captain Westoverheard an urgent radio conversation between theSEAFARER tow and the BALSA 37 altering theirpassing arrangement. Captain West and PilotWrasse could see the red sidelight and the rangelights of the BALSA 37. Pilot Wrasse observed thetug SEAFARER shudder. Both Wrasse and Westthen saw the range lights of the BALSA 37 close,then open as the ship turned to port. Its redsidelight disappeared and its green sidelight cameinto view.

Seconds later, the BALSA 37 and theSEAFARER tow collided. Captain Westimmediately sounded the danger signal andgeneral alarm, and backed the port engine

full. He gave no commands to the tug EDNA ST.PHILLIP. Struck on its starboard bow, theBALSA 37 yawed first to port, then abruptly tostarboard as the bow of the OCEAN 255 movedalong its side. The two vessels separated, andseconds later, the bow of the BALSA 37 impactedthe port side of the B. NO. 155 in way of its No. 1port cargo tank. Pilot Wrasse immediately made adistress call on Channel 16. Pilot Wrasse advisedCaptain West to make a course correction to keepthe tow in the channel. He then called the pilot boatMANATEE and directed them toward the scene topick up any survivors.

Following the two collisions, the EDNA ST.PHILIP initially let out 60 meters (200 feet) ofline to move away from possible hazards. Afterdetermining it was safe, the master of the EDNAST. PHILIP had his crew again make the tug fastalongside the B. NO. 155, where it remained untilbeing released at 1020. Due to its protectedlocation on the starboard side of the B. NO. 155,the EDNA ST. PHILIP was not damaged in thecollisions.

The BOUCHARD tow later anchoredapproximately one and one-half miles west of theSunshine Skyway Bridge.

The B. NO. 155 was later offloaded andtaken to drydock in Tampa, Florida fordamage survey and repairs.

f. Fire Fighting

Federal, state, county and municipal organizationsjoined forces in the firefighting efforts after CoastGuard Group St. Petersburg received the initial"Mayday" call at approximately 0548.

At 0615, 10 August 1993, the Tampa FireDepartment was notified of the vessel collisionsand subsequent fire aboard the OCEAN 255. Thecity of Tampa has responsibility for shipboardfires in Tampa Bay inside Egmont Key, includingthe channels leading to the Port of Tampa. Twofire department personnel were dispatched toCoast Guard Marine Safety Office Tampa tobegin planning firefighting efforts. The TampaFire Department's 12 meter (42 foot) boat FIRE 1was immediately dispatched.

Coast Guard vessels from Stations Cortez, St.Petersburg and Sand Key were dispatched to

14

search for survivors, assess vessel damage andassist with firefighting efforts. At about 0640, aCoast Guard helicopter confirmed that theOCEAN 255 was fully engulfed in flames near thesouthwest tip of Fort De Soto Park. By 0730, allthree Coast Guard small boats were applyingcooling water to the OCEAN 255. By 0745,Tampa Fire Department boats FIRE 1 and FIRE 2were on scene, along with the three Coast Guardboats, and were applying firefighting water.

Within two hours of the initial collision andignition of the OCEAN 255's cargo, the USCGCSITKINAK arrived as the Coast Guard's OnScene Commander (OSC). The SITKINAKcoordinated Coast Guard assistance to the vesselsinvolved in the collisions and fire. A four-manteam from the SITKINAK was dispatched to theBALSA 37, which was taking on water and had astarboard list. Flooding was controlled and theBALSA 37 stabilized when the vessel wasintentionally grounded east of Egmont Key. TheBOUCHARD tow reported no injuries, but the B.NO. 155 continued to release cargo.

Initially, the OCEAN 255 was ablaze from stem tostern and there was fire on the water surroundingthe vessel. Eventually the fire alongside the vesselburned out and the fires forward and aft of the No.6 starboard cargo tank were extinguished. The firein No. 6 starboard cargo tank presented the mostsignificant problem to firefighters because the16,098 barrels of Jet-A fuel were directly exposedto the atmosphere (the tank top having been blownoff by an explosion).

At 0830, the USCG Captain of the Port Tampaclosed the channel to all vessel traffic with a safetyzone established within three miles of the burningOCEAN 255. The main ship channel wascompletely reopened at 0630, 12 August 1993.

Maritrans officials contacted Gulfcoast TransitCorporation to ask for assistance in fighting thefire. When the casualty occurred, GulfcoastTransit was in the process of installing a 2500GPM firefighting system aboard the tugboatGULFCOAST. The installation was completedthat morning by removing two fire monitors from aformer Tampa Fire Department fire boat andinstalling them aboard the GULFCOAST. At 1215the GULFCOAST departed Gulfcoast Transit'sBig Bend facility with

two Tampa Fire Department firefightersaboard.

The Tampa Fire Department requested the ArmyReserve's 231st Transportation Company toprovide three landing craft (LCMs) to transportfire department equipment to the site. One 1000GPM fire truck with an aerial nozzle unit (EngineCompany 9) was loaded aboard one of the LCMsat Bayborough Harbor in St. Petersburg. Atractor trailer rig containing 69 drums (55 gallon)of AFFF was loaded aboard another LCM. Asmaller truck containing 24 drums of AFFF wasloaded aboard a third LCM. The process ofloading Tampa Fire Department equipment ontothe LCMs was not without difficulty in that thelong trucks would not roll easily onto the vessels.

Coast Guard, fire department and commercialresources continued to cool the OCEAN 255throughout the morning and afternoon of 10August. Firefighting efforts were temporarilysecured at 1650 due to intense thunderstorms, andresumed at 1800. The USCGC DECISIVEarrived on scene at 1830 and relieved USCGCSITKINAK as OSC.

By 2100, approximately 15 hours after the firebegan, the Tampa Fire Department was preparedto begin the primary assault on the cargo tank fire.The tug GULFCOAST and Engine Company 9aboard the LCM provided two foam streams witha combined application rate of 1900 GPM.

The primary fire was extinguished atapproximately 2145. At 2235, the Tampa FireDepartment OSC reported that the fire on boardthe OCEAN 255 was limited to one "hot spot" inthe forward port side.

At 0337 the OSC reported that there were still "hotpockets" on OCEAN 255. A reflash occurred at0347 and was extinguished by the Tampa FireDepartment boat on scene.

Tampa Fire Department officials reported that4,040 gallons of AFFF were expended theoverall firefighting effort.

. in

Just after daybreak on 11 August, personnel andequipment from Williams Fire Hazard & Control,Inc., Mauriceville, Texas arrived on scene. Theycontinued to maintain a foam blanket on theexposed cargo tank with

15

portable equipment which had been loaded on abarge.

Photographs and news articles indicate that bothCoast Guard and commercial crews fighting thefire on the OCEAN 255 may not have had properor adequate personal protective equipment.

g. Lifesaving Equipment and Rescue

On the morning of 10 August 1993, Mr. was the operator of the Tarnpa Bay Pilot's

Association 60-foot pilot boat EGMONT. The boatwas ferrying pilots to and from their assignedvessels. At approximately 0545, with theEGMONT approximately one half mile east of theEgmont Key Lighthouse, Mr. overheardurgent radio conversations on Channel 13. Helooked toward buoys 19 and 20 to see two meetingvessels. He then observed a fire and explosions, andproceeded rapidly to the scene to render assistance.

Upon reaching the burning SEAFARER tow, Mr. began searching for survivors. The water

was covered with gasoline and heavy oil. Hecoordinated his search with the Tampa Bay pilotboat MANATEE, which arrived shortly after theEGMONT.

Mr. a pilot boat operator, waslocated at the pilot station on Egmont Key. Atapproximately 0545, Mr. looked towardbuoys 19 and 20 and noticed a glow followedimmediately by an explosion and fireball. Mr.

placed the collision on a line between hisposition at the Egmont Key pilot station and thelights on the Mullet Key public fishing pier.

Mr. and his deckhand, Mr. ran to thepilot station pier and got underway in the pilotboat MANATEE. They quickly covered the 1.6miles to the SEAFARER tow which was engulfedin flames. Once on scene, they witnessed theexplosion of the No. 6 cargo tank.

The MANATEE initially located nine survivors bythe green chemical lights attached to their personalflotation devices (PFDs). Two additional survivorswere quickly located from the illumination of theirchemical lights. All survivors were recovered bythe MANATEE and accounted for. Atapproximately 0615, the survivors

were taken to the pilot station at Egmont Key towash the gasoline and other petroleum productsfrom their skin and clothing. The time from theinitial collision to the arrival at Egmont Key wasapproximately 30 minutes. Each survivor spentapproximately 15 to 20 minutes in the 87 F water.

All crewmembers from the SEAFARER donnedavailable U.S. Coast Guard approved Type IPFDs prior to abandoning the vessel. The vessel'screwmembers found the PFDs to be generallyeffective. Two survivors, however, found thewhistles attached to their PFDs had filled withfuel, making them unusable. One survivor said hecould not activate his chemical light as it waslocated on the inside of his PFD. It is not knownwhether the PFD was put on inside-out, or if thechemical light was attached to the inside of thejacket. Another chemical light broke duringactivation, rendering it ineffective. Testimonyindicated that the "Cyalume" chemical lights onthe PFDs were instrumental in locating thesurvivors in the predawn darkness.

Mate said that, while awaiting rescue,a wall of flame on the water's surface blewtowards the survivors. He considered removing hisPFD to dive under the approaching fire. As thewall of fire approached, a "break" in the flamesappeared which eventually became 8 to 9 meters(25 to 30 feet) wide. The survivors swam towardsthe break, and the fire passed them on both sides.

h. Injuries

Because of the warm water temperature andrelatively short exposure, none of the survivorssuffered from hypothermia.

Survivors from the SEAFARER tow were takento the Egmont Key pilot station, where theyrinsed petroleum products from their skin andclothes. Several complained that the gasoline andother petroleum products were burning their skinand causing irritation.

Mr. , a SEAFARER crewmember,received second degree burns to a four inchdiameter area of skin on his left forearm and wrist.He was treated at Tampa General Hospital at1350 on 10 August 1993 and released.

16

Other than Mr. , there were no injuriesreported aboard any of the other vessels involvedin the collisions. Additionally, there were noinjuries to personnel involved in the fire fightingeffort.

i. Drug Testing

BALSA 37: Pilot departed the BALSA 37at approximately 0830. He drove to St. Petersburgto a local clinic, but was refused admission withouta doctor's order. He then drove to Medi-Physicals,Inc., in Tampa. At 1030, he provided urine andblood samples for drug and alcohol testing. Thedrug tests were performed by Consulab/CedarsMedical Center, Miami, Florida. The test foralcohol was performed by the University ofFlorida, Diagnostic Ref. Lab, HSC, Gainesville,Florida. All tests were .

The BALSA 37 bridge crew departed the vesselat 1430, escorted by Coast Guard personnel toStation St. Petersburg. The chief mate,helmsman, and third mate provided blood andurine specimens to Medi-Physicals, Inc. between2220 and 2315. The tests for alchohol and drugswere performed by Consulab/Cedars MedicalCenter. All tests were

Captain remained on the vessel andwas tested on the afternoon of 10 August 1993 bya representative from MediPhysicals, Inc. Heprovided blood and urine samples atapproximately 1845. The alcohol and drug testswere performed by Consulab/Cedars MedicalCenter. All tests were

SEAFARER: The crew was sent to the TampaGeneral Hospital and tested at 1330 with eachcrewmember providing blood and urinespecimens. Drug testing was performed byNichols Institute, San Diego, California, andresults were reported by a medical review officeraffiliated with Greystone Health Services Corp.,La Mesa, California. Alcohol tests wereperformed by National Medical Services, Inc.,Willow Grove, Pennsylvania. All tests were

BOUCHARD: Captain remained onboard after the casualty. A representative fromExamination Management Service, Inc.,Tampa, boarded the tug, collected a urinesample and administered a breath test

for alcohol at 1930. Alcohol and drug testing wasconducted by Compuchem Laboratories, Inc.,Research Triangle Park, North Carolina. All testswere

Captain departed the CAPT. FREDBOUCHARD at approximately 0900. At 1200, heprovided a urine sample for drug testing to Medi-Physicals, Inc. The tests for drugs were conductedby Consulab/Cedars Medical Center. The testswere He did not provide a blood sampleand no alcohol testing was conducted. Captain

was unaware that alcohol testing wasrequired and did not request it.

All testing was conducted as required by 46 CFR16.240 with the exception of the alcohol testing ofCaptain There was no evidence of the useof alcohol or intoxication on the part of Captain

j. Personnel Histories

The Board decided to review personnel histories toassist in evaluating the process by which amariner's license is renewed. Also, several of theparties-in-interest submitted to the Board formalevidentiary requests for background informationon certain mariners involved in the collisions.Finally, a high level of interest in personnelhistories by the local media raised the issue ofbackground checks and casualty histories of themariners. The Board announced that it wouldobtain and examine various personnel documentsof Pilot Baggett, Captain Chapman, Mate

Captain West, and Pilot Wrasse.

Background investigations were completed,including National Criminal Information Centerbackground checks, state driving records, originaland renewal applications for Coast Guard licenses,Coast Guard civil penalty cases, Coast Guardsuspension and revocation proceedings, and theState of Florida's Department of Business andProfessional Regulation records for state pilots.

The background reviews revealed thefollowing:

-Two mariners had no arrests oractions against their licenses.

-One mariner had two arrests and no actionsagainst his license. His federal license wasproperly renewed.

-One mariner had a DWI and no actionsagainst his license. His federal license wasproperly renewed.

-One mariner had four DUIs, two refusalsfor alcohol testing (driving), two arrests,three federal mariner license suspensions,three federal civil penalties for negligentoperation of a vessel, and two state pilotlicense suspensions.

Pilot Wrasse's chief mate license, withpilotage endorsement, expired on 27February 1992.

k. Recording of Radio Communications

No decipherable recordings of Channel 13 VHFFM are known to exist. Coast Guard Group St.Petersburg was not recording Channel 13 the dayof the casualty. A recording of Channel 13 byGulfcoast Transit Corporation was examined byFederal Bureau of Investigation laboratories andrevealed no decipherable bridge-to-bridge radiotransmissions between the BALSA 37,SEAFARER and CAPT. FRED BOUCHARD forthe critical time period prior to the casualty.Gulfcoast Transit Corporation's receiver waslocated at their marine offices adjacent to theTampa Electric Corporation's Big Bend PowerPlant, however it was out of range of the collisionsite.

l. Marine Board Visits

On 12 August 1993, the Marine Board membersvisited the incident site during daylight hours,boarded the SEAFARER and OCEAN 255 forfamiliarization and later boarded the BALSA 37.On 15 August, Board members visited the B. NO.155 and the CAPT. FRED BOUCHARD. On 22August, Board members again visited the BALSA37 to obtain further information. On 24 August,the Board made a nighttime transit of the casualtysite on a Maritrans tow to view aids to navigation,observe tow running lights, and become familiarwith the channel and accident site at night.

17

CONCLUSIONS1. As the BALSA 37 passed abeam Buoy 21, nocourse change was ordered by the pilot. However,he hzd intended to order a course change to 262 T. Neither the helmsman nor the mate on watchheard or acknowledged such an order. Thehelmsman continued to steer his last orderedcourse of 261UT. This course, even with a currentdrift of 0.7 knots and set of 070 T. carried theBALSA 37 on a track further south in the channelthan intended by the pilot, who believed he hadordered a course of 262°T. Further, the pilot failedto determine his position in the channel afterpassing Buoy 21 and while transiting toward theturn in the vicinity of buoys 19 and 20. When thepilot observed a vessel apparently crossing hisbow, he was at a point approximately equidistantbetween, and just to the east of, buoys 19 and 20.However, he thought he was well to his starboard(north) side of the channel.

2. The SEAFARER tow was transitOing EgmontChannel on a course of 101 T with CaptainChapman at the helm. At the instant the BALSA37 pilot observed the green light, or lights, of theSEAFARER tow apparently crossing his bow, theSEAFARER tow was just to the starboard (south)side of the channel's centerline and had nearlycompleted passing the BOUCHARD tow. TheSEAFARER tow was at a point where it wouldhave to commence its turn to port into Mullet KeyChannel to avoid shallow water beyond Buoy 20and to avoid possible interference with theBOUCHARD tow making its turn. The testimonyof several witnesses, as well as analysis of thephysical evidence, suggests that the SEAFARERtow had commenced a gradual turn into MulletKey Channel as it approached buoys 19 and 20.Additionally, there was hydrodynamic interactionbetween the SEAFARER tow and theBOUCHARD tow as they passed. Thisinteraction, coupled with the application of leftrudder necessary to commence the turn, caused theSEAFARER tow to yaw to port. The SEAFARERtow thus presented a green light, or lights, to thepilot on the outbound BALSA 37.

3. BALSA 37 and the SEAFARER tow collidedat a position approximately equidistant betweenbuoys 19 and 20 and on, or just east of, a linedrawn between these

18

buoys. This position was best determined by twoTampa Bay pilot boat operators. One pilot boatoperator at the Egmont Key pilot station placedthe collision on a line between the station and thelights of the Fort De Soto pier. A secondoperator, standing by in a pilot boat just east ofEgmont Key Lighthouse, placed the collisionbetween buoys 19 and 20. These operators, aboutto proceed to the scene to rescue survivors, wouldhave carefully determined the collision site andwere in a good position to do so. The BALSA 37and B. NO. 155 collided at a positionapproximately 91 meters (300 feet) northwest ofBuoy 20.

4. The time of collision between the BALSA 37and the SEAFARER tow was approximately0545. The collision between BALSA 37 and theB. NO. 155 occurred less than a minute after thefirst collision.

5. The situation between the BALSA 37 andSEAFARER tow was "in extremis" atapproximately one minute prior to collision.Concern was voiced by both vessels atapproximately 30 to 45 seconds prior to collision.

6. The angle of impact between the centerlineof the BALSA 37 and the centerline of theSEAFARER tow was approximately 25degrees. The angle of impact between thecenterline of the BALSA 37 and thecenterline of the BOUCHARD tow wasapproximately 40 degrees.

7. The decision by Captain Chapman to overtakethe BOUCHARD tow as he approached buoys 19and 20 in Egmont Channel was imprudent andunnecessary. It created a situation where two largetows and a deep-draft vessel would meet at or neara turn at night. Further, there was an approximatethree knot speed differential between the two tows.By not overtaking, the SEAFARER tow wouldhave been delayed only about one-half hour inreaching its final destination. Also, it wasimprudent of Pilot Wrasse and Captain West toassent to the overtaking maneuver under thesecircumstances.

8. Once the overtaking commenced, all pilotsand operators should have carefully

monitored the situation as the three vesselsapproached the turn at buoys 19 and 20. Further,assuming the situation was being monitored, itwas incumbent upon each mariner to maintainclose communication regarding their positions andintentions. When it appeared BALSA 37 wasapproaching the turn at about mid-channel, doubtor alamm should have been raised well before the"in extremis" situation developed.

9. The apparent cause of this casualty was thefailure of the pilots and operators of all three ofthe involved vessels to adequately communicatetheir intentions and actions, or failure to query asto the intentions and actions of other vessels, inthat it was apparent they would meet at or nearthe turn at buoys 19 and 20.

10. There is evidence that Pilot Baggett:

-failed to effectively use the BALSA 37 bridgecrew and available visual and electronic aids tomonitor vessel activity in Tampa Bay in thevicinity of Egmont and Mullet Key Channels(Inland Navigation Rule 7 (a) and (b));

-failed to keep the BALSA 37 well to hisstarboard (north) side of Mullet KeyChannel, when it was in fact safe andpracticable to do so (Inland Navigation Rule9(a));

-failed to maintain a proper look-out by allavailable means so as to make a full appraisalof the situation and of the risk of collision(Inland Navigation Rule S);

-failed to adequately monitor Channel 13(Bridge-to-Bridge Radiotelephony Act);

-failed to recognize that more than two vesselsapproaching each other simultaneouslyconstituted "special circumstances" underInland Navigation Rule 2(a) (InlandNavigation Rule 2(a)); and

-failed to sound the danger signal as soon as hebecame in doubt as to the intentions of theSEAFARER tow (Inland Navigation Rule34(d)).

This matter has been referred to theCommander, Seventh Coast Guard District,

and the State of Florida Department ofBusiness and Professional Regulation forfurther investigation.

11. There is evidence that CaptainChapman:

-imprudently initiated an overtaking maneuverwhich resulted in three vessels meeting in theproximity of a turn in the channel;

-failed to advise BALSA 37 of theovertaking situation;

-failed to sound proper signals prior toovertaking the BOUCHARD tow (InlandNavigation Rule 34(c) and (h));

-failed to effectively use all available means,including radar, to monitor vessel activity inTampa Bay in the vicinity of Egmont andMullet Key Channels to determine risk ofcollision (Inland Navigation Rule 7 (a) and(b));

-failed to recognize that more than two vesselsapproaching each other simultaneouslyconstituted "special circumstances" underInland Navigation Rule 2(a) (InlandNavigation Rule 2(a));

-failed to question BALSA 37 on its delay ininitiating its starboard turn toward EgmontChannel; and

-failed to sound the danger signal as soon as hebecame in doubt as to the intentions of theBALSA 37 (Inland Navigation Rule 34(d)).

This matter has been referred to theCommander, Seventh Coast Guard District forfurther investigation.

12. There is evidence that Pilot Wrasse:

-imprudently assented to an overtakingsituation which resulted in three vesselsmeeting in the proximity of a turn in thechannel;

-failed to recognize that more than two vessels approachingeach other simultaneously constituted "specialcircumstances" under Inland Navigation Rule 2(a) (Inland Navigation Rule

2(a));

19

-failed to advise BALSA 37 of thedeveloping overtaking situation whenCaptain Chapman failed to do so;

-failed to effectively use all available means,including radar, to monitor vessel activity inTampa Bay in the vicinity of Egmont andMullet Key Channels to determine risk ofcollision (Inland Navigation Rule 7(a) and(b); and

-operated with an expired federal license.

This matter has been referred to theCommander, Seventh Coast Guard District forfurther investigation.

13. There is evidence that Captain West:

-imprudently assented to an overtakingsituation which resulted in three vesselsmeeting in the proximity of a turn in thechannel;

-failed to recognize that more than two vesselsapproaching each other simultaneouslyconstituted "special circumstances" underInland Navigation Rule 2(a) (InlandNavigation Rule 2(a));

-failed to advise BALSA 37 of thedeveloping overtaking situation; and

-failed to effectively use all available means,including radar, to monitor vessel activity inTampa Bay in the vicinity of Egmont andMullet Key Channels to determine risk ofcollision (Inland Navigation Rule 7 (a) and(b)).

This matter has been referred to theCommander, Seventh Coast Guard District forfurther investigation.

14. There is evidence of violation of 46 CFR 4.05-1(b) in that the CAPT. FRED BOUCHARD failedto notify the USCG Captain of the Port of thefailure of its starboard engine. This matter hasbeen referred to the Commander, Seventh CoastGuard District for investigation

15. There is evidence of inattentiveness on thepart of Pilot Baggett for several minutes prior tothe casualty. As he approached the turn, heshould have devoted his complete attention to theapproaching traffic. Pilot Baggett apparentlyallowed himself to be distracted by other matters,including pilot

dispatching arrangements. As a result, he failed tokeep himself fully cognizant of the approachingtraffic and did not recognize that the SEAFARERtow was overtaking the BOUCHARD tow. PilotBaggett also did not appear to realize the BALSA37 was proceeding at approximately mid-channelinstead of keeping to his starboard (north) side ofthe channel.

16. There is evidence that Pilot Baggett wasallowed to renew his federal master's license, withpilotage endorsement, with multiple DUI andcriminal convictions on his record. This matter hasbeen referred to Commandant (G-MVP) forfurther investigation.

17. Neither the Coast Guard nor the State ofFlorida licensing regimes presently appearcapable of systematically and reliably identifyingmariners who, by their actions professionally andin private life, should no longer be entrusted withthe duties and responsibilities of their licenses.

18. The actions of and , of the Tampa Bay pilot boatsMANATEE and EGMON1s, with regard torescue of the SEAFARER crewmembers arecommendatory and worthy of recognition. Thismatter has been referred to the Commander,Seventh Coast Guard District for consideration.

19. There is no evidence that drugs or alcoholcontributed to this casualty. All personneldetermined to be directly involved in the collision,as defined by 46 CFR Part 4, were tested foralcohol and drugs as required by 46 CFR 16.240,with the exception of Pilot Wrasse, who wasunaware of the requirement for alcohol testing andwas tested for drugs only (although there was noevidence of intoxication of this mariner). All testswere conducted within the required 24 hours andwere negative.

20. Several of the mariners and marine employersinvolved in this casualty were unfamiliar with thespecific requirements for mandatory drug andalcohol testing following a serious marineincident. Difficulties were encountered locatingfacilities which could perform the proper tests,and numerous phone calls and contacts had to bemade. In one case, two clinics were visited beforesampling was conducted.

20

21. Lifesaving equipment worked effectively andcontributed to the successful rescue of thesurvivors from the SEAFARER tow.

22. On the SEAFARER tow, the "InterconSystem" is equipped with a "high speed" and "lowspeed" mode to disengage the rams and disconnectthe tug and barge. Captain Chapman testified heattempted to use the "high speed" mode whichshould have taken about 10 seconds fordisconnection. However, in the excitement of theemergency, it is possible he used the "low speed"mode, which takes about two minutes todisconnect. Subsequent to the casualty, tests wererun on the connection system and it functionedproperly. Its design is such that it should havereleased under the loads experienced during thecasualty. Marking of the "high speed" mode alsoas "emergency release", or similar wording, mighthave eliminated any confusion.

23. Recording the critical radio transmissionsmade on Channel 13 in the minutes leading up tothe casualty would have assisted in thisinvestigation. Specifically, had Channel 13 beenrecorded to correlate transmissions to the precisetime of day, the process of reconstructing eventsleading up to the casualty would have beenenhanced.

24.The actions of the Tampa Fire Department,Gulfcoast Transit, Maritrans, and the otherfederal, state, and local agencies involved in thisfirefighting effort are commendable. Thisconflagration, requiring the transport of land-basedequipment to a relatively remote waterway site,was fought professionally. The time, however,required to place sufficient equipment on scene andto extinguish the fire was over 15 hours. A damagesurvey of the OCEAN 255 and interviews ofMaritrans employees indicated that, had the firecontinued much past the time of extinguishment, amajor structural failure was likely, with a possibletotal loss of both tug and barge, and release of alarge quantity of cargo.

25. This casualty is typical of many other marinefires in which shoreside organizations struggle tobring sufficient, proper equipment and platforms tooften remote and difficult to access waterside sites.The shipboard principle of detecting and rapidlyextinguishing fires at their source is virtually

21

impossible to achieve for such majorconflagrations.

26. Communications during the firefightingoperation were a problem, as is frequently the casein such large operations involving many agencies.Radios of the correct frequency apparently werelacking such that agencies could not rapidly andeffectively communicate. Also, sufficient batteriesto maintain available communications during theprolonged incident response were lacking.

27. There is evidence that commercial crews andCoast Guard boat crews may have fought this firewith improper or insufficient personal protectiveequipment. The danger of explosion, flashback, orrelease to the immediate area of large quantities offlaming product was ever present. This matter hasbeen referred to the Commander, Seventh CoastGuard District for review.

28. Subsequent to the collisions, cargo wasreleased from tanks on the OCEAN 255 and theB. NO. 155. The breach of tanks on the OCEAN255 was minor resulting in a very small cargorelease. A double hull design would likely haveprevented such a release. The port bow of the B.NO. 155 sustained major damage, including abreach of a cargo tank resulting in a large loss ofcargo. It is unlikely that a double hull on the B.NO. 155 would have significantly mitigated therelease of product following an impact of suchmagnitude.

29. A VTS employing the latest technology mighthave assisted in the prevention of this casualty. Itcould have provided information as to the generalvessel traffic situation including when, where andthe type of vessels that would be met duringtransit. A VTS could have determined that thethree vessels would meet at the turn and that theBALSA 37 was proceeding down the center of thechannel as it approached the turn. Radiotransmissions could have been monitored, raisingthe question as to the BALSA 37 pilot's awarenessof the overtaking situation. Finally, a VTS couldhave initiated a call for emergency response at theinstant it became apparent the collision wasimminent.

30. It appears that the chief mate on the bridge ofthe BALSA 37 at the time of collision was notactively participating in the

navigation of the vessel. Further, he did not appearto be cognizant of the vessel's precise position inthe channel nor the positions of the oncomingvessels. He perceived his duty as solely operatingthe engine order telegraph. The pilot's role is to actin an advisory capacity to the master, in this case,the chief mate as the master's representative. OnBALSA 37, Pilot Baggett seemed to be in totalcharge of the vessel's navigation, with virtually nocheck on his actions by the vessel crew, and withno interaction between the pilot and mate.

31. There is no evidence that mechanical ormaterial failure, or design deficiency, contributedto this casualty.

32. The cause of this casualty is attributable solelyto human factors. However, there is no evidencethat health, fatigue or the knowledge and ability touse equipment were factors. All mariners wereacquainted with the waterway. None of the vesselswere difficult to maneuver or underpowered, andthe shiphandlers knew the maneuveringcharacteristics. With the exception of the BALSA'spilot, the mariners were not

distracted and were aware of the speed andpositions of their own and the other vessels. TheSEAFARER's master made the decision toovertake such that the three vessels would meet atthe turn. However, none of the others raisedconcern over such a meeting.

It appears that complacency may have been afactor. It may have been the attitude of themariners that all would be well and to hope for thebest. Each knew the other vessel was in the handsof a mariner who was familiar with the waterwayand had made many transits. There was noweather, or other factor, which would raiseconcerns. Simply, the standard of care and level ofawareness was not what it should have been.

33. Except as noted above, there is no evidence ofactionable misconduct, inattention to duty,negligence or willful violation of law or regulationon the part of licensed or documented persons, norevidence that the use of drugs or alcohol, norevidence that failure of inspected material orequipment, nor evidence that any personnel of theCoast Guard, or any other government agency orany other person, contributed to the casualty.

RECOMMENDATIONS

1. That Commandant develop and implement amerchant mariner's license and documentrevocation procedure based upon a history ofconfirmed improper actions and/or violations. Asystem of points should be assessed for violationsof the Code of Federal Regulations, provedfindings of charges in suspension and revocationproceedings, and civil and criminal convictions.When the accumulated points exceed a specifiedlevel, the Coast Guard should automatically initiaterevocation proceedings. It is further recommendedthat the State of Florida develop a similar systemfor pilots under its jurisdiction.

2. That Commandant develop a new license anddocument application and renewal procedurewhich will determine and list all prior suspensionand revocation actions, and other criminalconvictions and DUI convictions. If additionalstatutory authority is necessary to perform suchchecks, it should be sought.

3. That the Tampa Bay Pilots Associationeducate member pilots on the requirements forpost-accident chemical testing.

4. That the Tampa Bay Pilots Associationcontract with a qualified facility to expedite thecollection of samples for drug testing itsmembers on a 24-hour basis.

5. That the Marine Safety Office, TampaMaHne Firefighting Contingency Plan bereviewed and revised as necessary to improvecommunications during such emergencies.

6. That, nationwide, Commandant establish meansfor recording Channel 13, or the designatedbridge-to-bridge channel, in critical waterwayswhere such recording is not currently being doneby a VTS or other system. Such recordingsfrequently are of great value in casualtyinvestigations.

23

7. That Commandant explore alternative oradditional noise-making devices on PersonalFlotation Devices whose operation would beunaffected by oil or chemicals on the water.

8. That Commandant initiate a review of themarine firefighting capability in the nation's ports,on its waterways, and offshore. As a result of theEXXON VALDEZ grounding, vast resources havebeen devoted to pollution prevention and response,including contingency planning, drills, andresponse hardware. However, an effective andrapid firefighting capability to prevent the releaseof pollutants through explosion and/or heatinducedstructural failure should be considered an integralpart of the pollution prevention and responseequation. Included in a national review should bethe consideration of the use of portable, highcapacity foam equipment. Such equipment wasprovided for this casualty, however, the Texas-based contractor arrived after the fire wasextinguished. The equipment was employed tomaintain a foam blanket on the exposed tank. Thisequipment has been used successfully to fightseveral marine petroleum cargo fires. It is highlyportable, relatively easy to operate and can deliverlarge quantities of foam and dry chemicalextinguishing agents. Such equipment could beeasily deployed to the scene of a fire on a varietyof floating platforms.

Monitors, prime movers, and foam stocks couldbe strategically located in port areas andtransported by land, sea, or air to areas within aport, to remote waterways sites, or offshore.Within certain small ports, such equipmentcould substitute for fireboats, precluding amajor capital investment.

Readily deployable, high capacity, portablefirefighting equipment could be the decidingfactor in preventing or minimizing the furtherrelease of pollutants to the environment in anynumber of casualty scenarios.