Historic Shipwrecks: Science, History, and Engineering
Searching for the Bonhomme Richard
About the Presenter:
|Melissa Ryan is the Project Manager at the Ocean Technology Foundation, based in Mystic, CT. She manages all aspects of the search for the Bonhomme Richard, including at-sea operations, international relations, logistics, research coordination, and development efforts. She has participated in nine oceanographic research and exploration missions ranging from marine archaeological surveys to the investigation of deep-water corals, and served as Chief Scientist on four of them. Her undersea experience includes three submersible dives, and five days spent aboard the USN Submarine NR1. She is also the Lead Program Instructor for NOAA's Office of Ocean Exploration and Research, and teaches professional development workshops for educators around the country.|
Compare the circumstances and technologies of the Bonhomme Richard surveys to date
Explain the integration of history and technology in archaeological surveying for an 18th-century shipwreck
Compare the log entries of an expedition aboard a modern research vessel, a submarine, and the Bonhomme Richard
Throughout the semester, we’ve presented lectures on various aspects of shipwrecks – from the historical components, to the technologies used to relocate, identify and conserve them, to the application of computer technology used in relocating those that have disappeared. Few ships of the 18th century hold as high a degree of historic significance as the Bonhomme Richard (BHR) (Figure 1).
Figure 1. The Bonhomme
Image courtesy of William Gilkerson.
Commanded by John Paul Jones, the BHR sank after defeating the British ship HMS Serapis off Flamborough Head, England during the American Revolution. It was during the Battle of Flamborough Head that his famous words, “I have not yet begun to fight!” became part of American history. Jones’ victory was a turning point in the Revolutionary War because it showed the world that the young Continental Navy was a formidable force, and strengthened vital French support for the War. Finding the remains of this warship would rekindle public enthusiasm for America’s naval heritage, allow for further studies on the construction of Revolutionary War vessels, and present the only means for interpreting the lives of the BHR crew members, of which little is known.
In 2005, the non-profit Ocean Technology Foundation (OTF) and the Naval History and Heritage Command partnered to conduct a multi-year initiative to locate the remains of the BHR. Project objectives are to:
Consolidate all archival,
historical and physical information known about the ship and battle
Construct search probability
matrices using Geographic Information System mapping technologies
Interpret the success of computer
drift modeling technologies as applied to archaeological surveying
Conduct a comparison of marine
surveying technologies and their specific applications to the search
Promote awareness and appreciation in students, educators and the public of Captain John Paul Jones, the Battle of Flamborough Head, and its historical significance
Over the course of thirty-five years, the project's historical researcher, Peter Reaveley, performed numerous literature searches and combed archives and technical libraries around the world to identify all available data relating to the BHR’s construction, operations, and loss. This historical research included the collection, interpretation, cross-referencing and correlation of:
A detailed timeline was developed which spanned the battle to 48 hours afterward, integrating all available and most relevant data for use in a computerized drift model, which we've already heard about in Professor Guth's lecture earlier in the course. An excerpt of the timeline from Friday, September 24, 1779 appears below. The battle occurred on Thursday evening, September 23, and the BHR sank on Saturday morning, September 25th. The vessel Alliance, referenced below, was a ship in Jones' fleet and was commanded by Captain Landais.
|Friday, September 24, 1779||0800||Fanning (midshipman on BHR): “We were now (anchored during the battle) something more than a league EbS True (079 True) from a point of land called Flamborough Head”|
|Friday||0900||Landais on Alliance: at 9 o’clock…came up with BHR...at 10 o'clock the Bonhomme Richard coming right up on the Alliance almost collided.|
|Friday||0900||"At 9 am the French ships were seen from Flamborough Head at the ESE" (112°True)|
|Friday||1000||"At 10 am the fleet was laying ESE (112°True) from Flamborough Head distance 2 leagues." "…six sail were seen about 2 leagues off at sea, much shattered…they kept their station all that day…"|
|Friday||0900-1000||An eyewitness at Flamborough Head described the ships as laying-to ESE (112˚T.) off Flamborough Head, two leagues off (6 nm), between 9 a.m. and 10:00 a.m. on Friday morning. By that time the sea-fog had begun to clear and visibility had improved.|
|Friday||1000||At 10 a.m. the fleet was laying ESE (112°True) from the Head, distance 2 leagues.|
|Friday||1000||Alliance took Serapis in tow to draw her from the land...had drifted to within 4 miles of Flamborough Head...At 9:00 we came up with the BHR. At 10:00 the BHR coming right upon the Alliance, towed the Serapis by..." Alliance towing Serapis to the NE (045°True) or NNE (022°True)|
|Friday||1100||John Paul Jones: When fog cleared around 11:00, enemy convoy had taken refuge under Scarborough Castle|
|Friday||1200||After the battle, six sail of ships were seen from Scarborough, bearing from the castle ESE (112° True) distance 5 to 6 leagues..|
The timeline offers clues as to the locations of the ships and the crew actions at certain times, and can ultimately help in determining a search area.
Geographic Information System Mapping
OTF created a comprehensive Geographic Information System (GIS) map to visualize the project data geographically. Information about individual shipwrecks and reported obstructions off the coast of the United Kingdom (U.K.) was obtained through the U.K. Hydrographic Office Wrecks Information service, and was plotted in the GIS database (Figure 2). Each entry was hyperlinked to an information sheet which included the wreck’s location, water depth, scour, debris field, sonar contact length, surveying details, and general comments. The majority of these wrecks have been identified, but for some, very little information exists. The database also includes information on seabed composition and topography, pipelines, wellheads, previous search areas, and sonar and magnetometer contacts derived from previous surveys.
|Figure 2. A screen capture of the GIS map.
The southernmost peninsula is Flamborough Head. Each dot represents a
shipwreck, or known hazard to shipping.
Image: M. Ryan
This GIS resource raised survey efficiency by allowing search teams to bypass known wrecks and closely examine unidentified ones. The boundaries of the survey area were also entered into the map, which allowed the area to be correlated with locations of known wrecks and other data such as the historic sightings of Jones’ squadron. The map is an evolving product intended to establish new protocols for integrating GIS into marine archaeological surveying.
Computerized Drift Modeling
The BHR’s log and eyewitness accounts agree that the ship did not sink immediately after the battle, but drifted for approximately 36 hours before sinking. Computer modeling of the BHR’s drift after the battle provides a scientific and testable hypothesis for locating the wreck. A predictive model was developed for the location of the battle, the positions of the fleet after the battle, and the general location of the BHR’s sinking. This modeling effort localized the offshore locations of the battle and ships for the first time, and served as a useful tool in facilitating ongoing data analysis.
The drift simulations incorporated intensive archival research, integrating all available eyewitness and crew accounts of the battle into one chronological sequence and identifying the most probable positions of Jones’s fleet and BHR prior to its sinking. The Climatic Research Unit at the University of East Anglia prepared a historical weather analysis for the date of the battle, including daily synoptic weather charts, using historical data and eyewitness reports. A historical tidal current analysis was also prepared. All of this information was integrated into the drift model and a series of simulations were run using the variables of sail, wind and current. Data runs also considered known sail combinations flown at certain hours during the drift period as well as the increasing water levels in the struggling ship’s bilge. Multiple iterations of drift modeling gave insight into the credibility of some eyewitness accounts, reinforcing some and weakening others. This resulted in a predictive model which bracketed BHR’s possible drift patterns and identified a main search area for the location of the sinking. The drift model was merged into the GIS in order to interpret all data simultaneously during the survey.
The first search for the BHR took place in 2006 and was a three-week survey utilizing side scan sonar and a magnetometer. As a result of this survey, five shipwreck sites, or targets, were identified which warranted closer investigation during a subsequent expedition. The targets were prioritized based on their size, shape, and magnetic signatures, and appeared to be shipwrecks and/or man-made objects. Only one of the targets did not appear in the UKHO wreck database. Although the other targets were in the database, little or no information was known about them. All were worthy of investigation in that they may have been wrecks of historical significance to England or another country’s maritime history. These targets and their characteristics appear in the table below.
|Table 1. Shipwreck targets of interest from the 2006 Expedition.|
Expedition 2007 - ROV Troubles
In 2007, the team partnered with the Office of Naval Research to conduct a five-day expedition aboard a research vessel using a Remotely Operated Vehicle (Figure 3) to attempt to classify the targets located in 2006.
|Figure 3. The SeaEye Falcon ROV
on the deck of the RV Oceanus.
Photo: M. Ryan
Representatives from English Heritage, the U.K. organization that manages England’s historical sites, and Woods Hole Oceanographic Institution also participated in this expedition. Equipment problems plagued this mission, as noted in the following log excerpts. Note the amount of time that can be spent repairing the technology being used. Even with backup equipment, nothing is certain.
14 August 2007
00:04 ROV released. Speed of the current appears to be a problem for the ROV.
00:06 Estimated 1 KT current from the North Northwest using shipboard ADCP measurements. The depth is 65m.
00:15 Captain disengaged the stern prop to ensure the cable did not engage the prop.
00:20 Power breaker tripped and lost power to the ROV. Working to restore power.
00:25 Ship heading is 145°T and current is 1.15KT from 325°T.
00:26 Replaced power strip and restored power to ROV.
00:45 ROV back on deck. The plan is to use a depressor weight to get the cable down to the bottom to prevent the ROV from being dragged by the cable in the current.
01:25 ROV in the water.
01:45 Lowering depressor weight and cable.
01:55 Bottom in sight. Current .34KT. Ripples on the bottom. Cable readout shows 215f depth.
02:35 Current at .13KT Bottom in sight. Power breaker continues to trip. Deep ripples on the bottom.
03:02 ROV team request to reposition ship. Bottom current at .21KT to .02KT. Reposition ship with bow thruster.
03:32 Continue with repositioning of ship. Crew will recover anchor to reposition. Concern with ROV cable and anchor chain. ROV group on deck to recover ROV. ROV group predicts cable is wrapped around anchor chain.
04:08 ROV on surface
04:20 ROV cable wrapped 1.5 times around anchor chain. Launching small boat to untangle cable. Captain instructs crew to wake up anyone that is needed to assist.
05:11 ROV on deck. Cable intact and on deck.
05:20 Work boat on deck
05:30 ROV team working on electrical problem. ROV team testing cable.
07:11 ROV team still trouble shooting electrical problem. ROV on deck.
(Fast forward about 8 hours....)
15:00 Ship repositioning over target #1 (V0241) ROV (small) not communicating. ROV team trouble shooting problem.
15:52 Standing by while ROV group continues to repair and trouble shoot problem with smaller vehicle.
19:20 Continue to stand by for ROV group to trouble shoot larger ROV. Smaller ROV is non-operational. Plan is to deploy side scan. ROV group setting up side scan system.
19:20 Too rough to deploy side scan. Focus now on repair of larger ROV.
The strong currents overpowered the smaller backup ROV, which took on some seawater, and its electrical circuits were fried (Figures 4 and 5.)The main ROV did eventually become operational again, but time only allowed the investigation of Target 1 from 2006 and a new target, which was found to be a wellhead.
|Figure 4. The tiny backup ROV on the
2007 expedition, which at one point was dragged backward over a target
due to the strong North Sea currents.
Image: R. Neyland
|Figure 5. The backup ROV met its
demise after one dive. This image shows its circuit panel after
the current yanked its tether and created an electrical fire inside it.
Image: M. Ryan
Target 1 (see Table 1 above) appeared to be a debris field of large cut stone, and the project archaeologists hypothesized that it might have been a vessel carrying quarry material. Stones of various shapes littered the sea floor in its vicinity, but evidence of any iron, wood, or artifacts that could be associated with BHR were lacking. Target 2 could not be relocated on this mission, but a year later, the NR1 was able to investigate it (see below). After reviewing all the sonar and magnetic data, and based on comparisons with BHR's dimensions, structure, composition, and other factors, it was determined that none of the 2006 targets presented a compelling case for being the BHR.
Expedition 2008 - The NR1 Adventure
The U.S. Navy Submarine NR1, as we've heard from CDR John McGrath, was deactivated in 2009 and was the Navy’s small nuclear submarine that could be used for civilian research (Figure 6). It was available for civilian use during certain parts of its 40-year lifetime. Scientists could submit proposals for its use, and if the NR1 was operating in the same area as the scientists were exploring, if it had extra time between military missions, and if it did not get called away on a military emergency, the scientists were offered the opportunity of a lifetime.
|Figure 6. Submarine NR1 and her support ship MV Carolyn Chouest off the coast of Norway.|
The 2008 search for the BHR was the NR1’s final mission. For three weeks, operating 24/7, the submarine surveyed 400 miles of the ocean floor, using Obstacle Avoidance Sonar as its primary survey tool. The NR1 was the most efficient surveying technology used in any of the BHR searches to date. Operating independently of the temperamental North Sea weather, she investigated 26 shipwrecks with side-looking sonar and video documentation. Target 2 (from the table above) was investigated and found to be a 25 foot-long tangled mass of fishing nets. The team had no way of determining what the nets were hung on, and it was too dangerous for the submarine to get too close to them.
The excerpt from the NR1 log below details their
investigation of a wreck.
JUNE xx, 2009
Fix Reset: N/A Computer Reset: No
Ship is submerged, independent, depth 157 ft
Ship is hovering over wreck
Operating IAW CTF 69 Subnote S/N 07, DTG 201623Z, with Amend 1.
The XO is the OOD.
0050 Properly relieved by the ENG.
0230 Completed in depth visual and video surveys of wreck. Wreck is listed to port between 20-30 degrees. The measured width of the wreck at a distance 70 ft forward of the stern is 22 feet. Based on a 30-degree list, that would correlate to an actual width of 26 ft. Positioning the ship for visual surveys on the debris field off the bow of the ship, as well as an additional SLS pass. The ship is oriented with the bow pointing approximately 180.
0300 Obtained LF SLS (Side Looking Sonar) pass on wreck. File is saved as M8051993. SLS image shows the wreck to be approx 31 meters long, 4 meters high and 7.5 meters wide.
0316 Gained VSOB at 22 feet altitude. Bottom is flat and silty, with no rocks. There are no trawl marks. D/A sonar indicates a soft bottom with 2 ft silt covering. There is no vegetation and little marine life.
0328 Ship is bottomed at Nxxxxx Exxxxx
0341 Objects determined to be plant life on the bottom, not contacts of interest or evidence of a debris field.
0343 Lifted off the bottom. C/A 20-25 ft.
0424 Mark sunrise.
It's interesting if we think back to Peter Reaveley's lecture in Week 2 and the log of the Bonhomme Richard (Figure 7), and compare it to the log of NR1, and with those of an expedition aboard a research vessel. What elements are similar? How do the log entries change depending on the vessel and technologies available to the mission?
|Figure 7. Excerpt
from the log of the Bonhomme Richard. The original was
handwritten, and would have resembled
the log of the
Source: National Archives
The NR1 was a fantastic surveying technology and its sonars were so sensitive that they could detect an object the size of a paint can from nearly a mile away. As we've seen from the 2006 and 2007 expeditions, marine archaeology surveys are usually done in phases, mostly due to cost and the need for different working platforms to support the different technologies:
Use side scan sonar and magnetometer to locate targets of interest.
Phase 2: Return to targets to conduct visual investigations using a Remotely Operated Vehicle or, if depth allows, divers.
The NR1 was able to combine both phases into one efficient survey, using video for target investigation. Although it was not equipped with a magnetometer, its visual surveys were thorough and detailed, and its Obstacle Avoidance Sonar was able to detect some partially or mostly buried objects. Once a target was located on sonar, the ship could mark its location and return to that site to document it using video. When compared to the 2006 survey, the NR1 covered about 8 times more seabed in the same amount of time, and visually investigated 26 different shipwrecks. The wreck mentioned in the log above, Number 6008, is pictured below, along with other unidentified wrecks (Figures 8-11).
investigating a wreck, the NR1 detected a strong 30-meter-long contact on the OAS
about 300 meters away. Thinking it might be debris associated with the
wreck, the crew investigated, but when they arrived at where the target should
have been, they saw nothing except seabed. This target was detected several times, and the crew investigated each time, but never saw
This mystery would be solved in 2009.
|Figure 8. This wreck appears to be in three pieces, but the missing pieces are probably buried in the sediment.|
|Figure 9. Wreck mentioned in the log excerpt above.|
|Figure 10. A sonar image and a video composite show very different views of this wreck.|
|Figure 11. What happened to this
wreck? It looked like something had collided with it and pushed in its
side. The destruction was so massive that it was difficult to even
describe the wreckage, as is relayed in this excerpt from the NR1 log:
"1140: Completed LF SLS pass on wreck. File saved as M8052110. Approximate target dimensions are 59 meters long, 12.4 meters wide and 3.7 meters high in the center. The center of the vessel appears to be relatively intact, and the rest of the vessel (bow and stern) appear to be pretty low to the ground and may be damaged. Positioning for a visual investigation. Approximate orientation 150/330.
1304: Completed visual surveys of wreck. The wreck appears wooden, but there was no observation or any indications of metal, cannons or cannon balls. Visibility was very poor, but the portions observed were significantly destroyed such that visual surveys could barely recognize the contact as a shipwreck as opposed to a pile of timbers."
In winter 2008/2009 the Marine Technology Society Journal published a paper on the search for the BHR.
Expedition 2009 - A New Discovery?
2009, at the request of the U.S. Chief of Naval Operations, the French Chief of
Naval Operations invited three members of the project team aboard a French Navy
vessel for the fourth expedition in search of the BHR. This mission
was particularly significant because not only did it test new, state-of-the-art
survey technology, it was also the most ambitious multi-national collaboration
to date. Members from the U.S. project team were joined by representatives from
the U.K. Hydrographic Office and the French Hydrographic Office aboard a French
Navy minehunting vessel. This significant French involvement stemmed from the importance of
the BHR to both U.S. and French maritime heritage, since the BHR
was on loan to the United States from France at the time of her epic battle with
The primary search tool for the 2009 expedition was a magnetometer. With this technology, the team was able to search a 50-square-mile area of the seabed, identifying numerous magnetic signatures in the process. As we now know from previous lectures, this type of technology is important because the BHR had iron ballast, which we would expect to detect with a magnetometer. The team also had the benefit of an experimental Unmanned Underwater Vehicle (UUV) equipped with high resolution sonar to classify any targets that were found (Figure 12). However, it was not intended to be a search tool, and was only used to investigate targets that were already located.
Figure 12. The Unmanned Underwater Vehicle used in the 2009 search for the
Photo by M. Ryan
Three shipwreck sites, initially identified through their magnetic anomalies or during previous expeditions were investigated with the UUV. While two of the targets were determined to be modern wrecks, the third target remained an intriguing mystery – it appeared to be a mostly buried wooden shipwreck (Figures 13-16), with no magnetic signature, and did not appear in the UK Hydrographic Office database of wrecks. This was the "mystery target" detected by the NR1, but since it was buried with little relief, the crew could not see it through the video cameras of NR1. Its outline was only visible from a distance above it.
|Figure 13. The mystery wreck detected by NR1, now known by the project team as "Buried Wreck," was visible only from above. In this side scan sonar image taken on the French expedition, it is barely visible in the seabed.|
|Figure 14. This image of Buried Wreck was taken with the multi-beam sonar on the Unmanned Underwater Vehicle. It shows an area in red which may be the stern of the ship. It projects less than a meter off the seabed.|
|Figure 15. Here is another image of the wreck from a different angle, with its length depicted as 29.95 meters.|
|Figure 16. Mine hunting sonar produced this image of
All of these images show examples of the different types of data that can be collected and which provide different levels of detail. Even with these three technologies, the difficulty of identifying this wreck is evident.
|Video. A fourth technology, a ROV, was used in a short follow-up mission to try to further classify this wreck. The WMV format video clip to the left (opens in separate window) reveals (mostly toward the end) many piles of stones scattered throughout the wreck site. It is possible that this could be stone ballast from the wreck.|
The discovery of Buried Wreck presented thought-provoking questions for the
project team. Could this possibly be the BHR? Why wouldn't it have any
magnetic signature? Could it be another wreck of cultural or historical
significance? Should further resources be expended to investigate it? What
other technologies would be helpful in classifying it? Later that summer, the
French Navy investigated this wreck with a ROV, which did not reveal much
additional information. Since the wreck was mostly buried, they saw only the
piles of stones on the seabed, and no apparent artifacts. The survey team
ultimately labeled it a low probability of being the BHR.
Expedition 2010 - USNS Henson
As we've seen throughout the course, searching for a specific shipwreck without the benefit of a somewhat reliable location of its sinking is extremely challenging. In September 2010, our survey team returned to the North Sea with four midshipmen to expand our search area with the use of two Autonomous Underwater Vehicles, multibeam bathymetry, and a towed side scan sonar. We noted 35 targets that we assigned an "A" priority, and were able to investigate a few of these in cooperation with the French Navy in April 2011. We had the use of a ROV, and collected some excellent video footage, but the sites still need further investigation with different technologies in order to clearly assess their characteristics.
|Figure 17. Side scan mosaic, at 1 m resolution, of part of the search area. This shows a number of NNW--SSW trending bedrock ridges with very strong (white reflections), and NE-SW trending ripples that are perpendicular to the tidal flow.|
|Figure 18. Multibeam bathymetry mosaic of part of the search area.|
|Figure 19. Full resolution sidescan image (resolution about 0.4 m) of a wreck imaged in 2010. Size, shape, and preservation all fail to match what we expect for BHR.|
History belongs to everyone, and it is important to the project team to share
this project with the public as much as possible, without compromising our years
of research and significant efforts to date. The Bonhomme Richard Project
represents much more than a search for a shipwreck. It serves to inspire people
young and old to become aware of our maritime heritage, and to be proud of it,
because it has helped to shape who we are today as a nation.
Return to course syllabus
Last revision 8/5/2015