5.1 Introduction

In 2010, the United States Congress appropriated one million dollars to identify the most ecologically and economically significant potentially polluting wrecks in US waters, a joint initiative between the U.S. Coast Guard and its Regional Response Teams, along with NOAA, through its Office of National Marine Sanctuaries, Office of Response and Restoration, and Maritime Heritage Program. Of some 20,000 US wrecks that postdated 1891, when most steam and motor vessels converted from coal to oil, the study focused on: vessels built of iron or steel (wooden wrecks would have deteriorated and already released their oil), cargo vessels over 1000 tons (smaller vessels would have small amounts of fuel), and any tank vessel. This resulted in a group of 600–1000 vessels.

Detailed research followed, working from available wreck reports, archival documents, and resulted in a focused desktop study of 87 vessels that were thought to pose a potential pollution threat due to the nature in which they sank, the amount of fuel or cargo likely still inside the wrecks, and the structural reduction and demolition of those that were navigational hazards. To further screen and prioritise these vessels, risk factors and scores were applied to elements such as the amount of oil that could be on board and the potential ecological or environmental impact. The contractors did the modeling forecasts, as well as ecological and environmental ‘resources at risk’ assessments. In nearly every evaluation of the 87 vessels, little to no details of the actual wreck as a physical entity was available that offered a forensic sense of the wreck, and few of the wrecks had even yet been located.

The NOAA study had a budget of one million (USD) set by Congress; equally divided, that means that each of the 87 studies would have cost $11,494.25. Actual costs varied, of course, and additional resources were brought to bear, such as NOAA and US Coast Guard salaries, but in the larger scale of government spending, the Congressionally set budget was minimal. Within that context, a national effort to locate and conduct a physical assessment of each of the 87 wrecks, an approach which had not been funded, was beyond practical reach. While some wrecks were in shallow water, and more readily accessible, others lay deep, where the technological means to reach them for assessment was cumulatively cost-prohibitive.

The obvious solution for a ‘next step’ that took the project from desktop to dive would come with oceanographic cruises, surveys and remotely operated vehicle dives of opportunity. Three separate opportunities in 2013, 2015 and 2016 resulted in physical assessments of the wrecks of SS Fernstream (1952), SS Coast Trader (1942) and USNS Mission San Miguel (1957) and subsequent revisions to the initial evaluations of their potential to pollute. The author was one of the NOAA managers involved in the NOAA PPW study and was principal archaeological investigator for two of the reassessments as Director of NOAA’s Maritime Heritage Program in the Office of National Marine Sanctuaries.

5.2 SS Fernstream

The Norwegian steel-hulled motor vessel Fernstream was built at Eriksbergs Mekaniska Verkstads Aktiebolag in Gothenburg, Sweden as one of six 8400-ton freighters built to the same plan in 1948–1949. Launched in July 1949, the 416 ft (126 m) long freighter entered service in October as a general-purpose cargo ship with refrigerated holds and accommodations for passengers. The Norwegian shipping firm A/S Glittre owned Fernstream, but it was managed for them by the Oslo-based firm of Fearnley & Eger under the Norwegian flag. Unlike earlier steamships, Fernstream and its sister ships were diesel-powered. Under Fernley & Eger’s flag, Fernstream and its sisters shipped cargo from the United States to Asia, and in some cases, immigrants to Asia Pacific countries.

On its final voyage, Fernstream departed San Francisco for Manila with 11 passengers, 3000 tons (2721 mt) of soybeans in bulk, and general cargo that included machinery for a hydro-electric power plant in the Philippines on December 11, 1952. It was foggy, and as Fernstream was in position to motor out of the Golden Gate, the inbound cargo freighter Hawaiian Rancher, an 8353-ton vessel, struck Fernstream on the port side near the bridge, penetrating the hull and flooding the after part of the engine room. The collision damaged the watertight bulkhead connecting the engine room to the #4 hold, and the crew was unable to close the watertight doors in the shaft alley. Flooding rapidly, and with a total loss of power, Fernstream sank quickly as the damaged but still afloat Hawaiian Rancher lowered lifeboats and rescued Fernstream’s passengers and crew.

Fernstream’s wreck was not salvaged nor was there any historical record of efforts to clear or reduce the profile of the wreck. When evaluated during the NOAA PPW study, Fernstream was classified as a moderate/high risk of polluting San Francisco Bay and the surrounding coastline, depending on tides and the transfer of water in and out of the Golden Gate, because there were no accounts of clearance, or of oil release at the time of the sinking. However, the study noted that there were no detailed reports of the sinking, other than local news headlines, and those focused on the human drama of the rescue of the passengers and crew. It was also assumed that the wreck was likely not broken up and was in one contiguous piece:

The Fernstream is classified as High Risk for oiling probability for shoreline ecological resources for the WCD because 100% of the model runs resulted in shorelines affected above the threshold of 100 g/m2. It is classified as Medium Risk for degree of oiling because the mean weighted length of shoreline contaminated was 52 miles. The Fernstream is classified as High Risk for oiling probability to shoreline ecological resources for the Most Probable Discharge because 100% of the model runs resulted in shorelines affected above the threshold of 100 g/m2. It is classified as Medium Risk for degree of oiling because the mean weighted length of shoreline contaminated was 18 miles (NOAA, 2013b: 27).

The prominent location of the wreck, just off Alcatraz in one of the most scenic harbors of the United States, while not a factor in the assessment of potential risk, did result in NOAA recommending further assessment of the wreck, using surveys of opportunity to locate and assess it.

The opportunity to do so came just as the PPW study was released by NOAA. A multi-year study of maritime heritage resources in Greater Farallones National Marine Sanctuary, headquartered in San Francisco, involved multiple partners (Delgado et al., 2020). In May 2013, NOAA’s Office of Coast Survey’s locally based Navigation Response Team (NRT 6) conducted two side-scan sonar surveys that relocated and mapped the wreck. The demonstration of a CODA Octopus Echo-Scope sonar system for the dive team of the San Francisco Fire Department’s Marine Unit in November 2013 also assessed Fernstream on NOAA’s recommendation. The sonar data from the three surveys was highly illustrative. Fernstream sits upright with the forecastle, partial remains of the stern-house and bridge-house, as well as the forward and after mast house structures visible. The hull has suffered from catastrophic collapse of the bridge house structure. The hull is breached on the starboard side forward of the bridge-house, the location of the collision. Masts, booms, and king posts have collapsed onto the deck; and the port side of the wreck is buried deep in sediment.

The stern is the highest remaining portion of the wreck above the seafloor but was seen to be partially collapsed. The bow has twisted to port, and the bow is no longer in longitudinal alignment with the after part of the hull forward of the bridge house. Outside the wreck, the starboard side of the hull shows evidence of sediment scouring, more prominently in the bow, a typical occurrence with shipwrecks due to their position on the seafloor in prevailing currents. Rather than a contiguous wreck, Fernstream is broken, and heavily shrouded in the thick muds of the bay. The deep tanks, where fuel oil was bunkered, are at the lowest levels of the hull, all of which is now buried deeply in bay mud.

Based on the sonar imagery, NOAA revised the PPW assessment for Fernstream, noting that the sonar indicated that Hawaiian Rancher appears to have impacted more than one deep tank containing fuel oil, and either from settling on the seabed or after decades of corrosion in the active currents of San Francisco Bay near the harbor entrance, Fernstream’s starboard hull has suffered a catastrophic collapse that likely also collapsed the #3 deep tank. The hull collapsed and is open to the sea where exposed, and likely no longer contains any significant amounts of oil, but the possibility of oil in sediment filled areas of the wreck remains a probability. The risk of the wreck polluting was modified from high to medium, with a recommendation for active monitoring if sediment movement released oil at a future date.

5.3 USNS Mission San Miguel

The USNS Mission San Miguel was a 524 ft (160 m) long T2-SE-A2 tanker, with a welded steel hull and turbo-electric propulsion. It was one of more than 500 mass-produced T2 tankers built to the design specifications of the U.S. Maritime Commission during the Second World War (see Brennan et al., Chap. 9, this volume). Built in 1943–1944 at the Marinship Yard in Sausalito, California for the War Shipping Administration, it was chartered to private owners until 1946, when it was laid up at the National Defense Reserve Fleet moorage near Mobile, Alabama. Acquired by the U.S. Navy in 1947 and commissioned as USS Mission San Miguel (AO-129) it was chartered again to private owners, before being transferred to the Military Sea Transportation Service as USNS Mission San Miguel. The tanker passed in and out of service with extended layups in reserve fleets through 1957, when it wrecked on October 8, 1957, after leaving Guam and heading across the Pacific. It ran aground during rain squalls on Maro Reef in the Northwestern Hawaiian Islands. Fully ballasted with sea water, it remained intact, and the crew was rescued (NOAA, 2013c).

The historical record, which includes US Navy salvage reports and damage assessment from the unsuccessful attempt to recover the vessel at the time of its wrecking on Maro Reef, along with a detailed assessment of the record by archaeologist Hans Van Tilburg, makes it clear that the tanker was in ballast at the time of loss, with no cargo (fuel oil and gasoline) on board (Van Tilburg, 2003). USNS Mission San Miguel was fueled and after running for 5 days had consumed an estimated 1700 barrels of Navy Special Fuel (No. 5) out of a 14,700-barrel capacity. The fuel oil was carried in two tanks, which were located aft of the engineering compartment bulkhead.

The tanker could not be freed and was abandoned to break up in that remote archipelago of atolls and reefs. The area of the wreck is now within Papahānaumokuākea Marine National Monument, primarily administered by NOAA in partnership with other agencies. The presence of the wreck in the marine protected area led to an assessment of potential risk in the 2013 study even though the wreck had yet to be located (Van Tilburg, 2002, 2003). Given the uncertainty of how much of the tanker’s fuel remained inside the wreck, it was rated a High/Medium, which at the time was noted as a conservative assessment reflecting the lack of information on both exact location and structural integrity. As annual Rapid Ecological Assessment and Monitoring (RAMP) cruises included maritime heritage components, the hope was that one of these cruises would locate and document Mission San Miguel.

That opportunity came with a RAMP cruise in August 2015. The wreck was located on August 3, 2015, and a preliminary archaeological assessment was undertaken by the team of archaeologists and scientists aboard the NOAA vessel Hi’ialakai. The vessel lies close to the reported area of its loss, in 80 ft of water off Maro Reef. An archaeological team documented the site and prepared a site drawing and a site report (Raupp, 2015) The 2015 assessment focused on 2 days of dives to document the characteristics of the site and compare those data with plans of USNS Mission San Miguel to identify the wreck and to assess its archaeological integrity as a site potentially eligible for listing in the National Register of Historic Places.

The wreck lies at a bearing of 080° and consists of three main sections. These include the largely intact stern section of the ship, which rests on its port side and rises from approximately 80 ft (24 m) at the seabed to a depth of 29 ft (8.8 m) below the water surface; the aft cargo tank section, which is located directly forward of the stern and is mostly collapsed; and an area thought to represent the tanker’s bow located approximately 540 ft (165 m) east and northeast of the stern section that includes two anchors and chain, numerous components of deck machinery, and scattered sections of hull plate and piping in depths ranging from 5 to 35 ft (1.5–10.5 m). Each of these aspects of the vessel’s deposition closely matches historical data pertaining to Mission San Miguel. According to US Navy reports, when salvage operations were terminated the tanker was aground with a bearing of 075° and drew 4 ft in the forward part and 83 ft in the after portion (Raupp, 2015).

The 2015 survey documented the disintegration of the majority of the hull of the vessel and that this had extended into the engineering spaces in the stern. As the vessel sank and moved deeper, it is likely that the tanker twisted and broke apart. The bow, being the shallowest portion of the wreck, has collapsed and broken apart. Only the stern is intact. Structural collapse of all but approximately 100 ft (33 m) or a fifth of the hull includes the fuel oil tank on the starboard side of the stern. About 18 ft (5.5 m) of the port side and therefore the port fuel oil tank is buried in the sand bottom. The archaeological team believes that, while this tank may not be collapsed, it may have opened up to some extent due to the pressure of the wreck. They observed no oil or sheening on the site. Divers did not enter the wreck but visible overhead spaces did not contain any obvious oil (Raupp, 2015).

When assessing the archaeological report, NOAA revised the risk assessment for the wreck. The Pollution Potential score for the vessel is now LOW for worst case discharge and LOW for most probable discharge. This downgrade was based on the change in knowledge of the wreck site, particularly the lack of remaining structural integrity in the areas of the wreck that would have held fuel oil. No further work was recommended as the only other archaeological work that could be undertaken would be excavation of the buried port tank area but that could have released trapped oil. The existence of oil residue or trapped oil beneath portions of the intact stern is possible, but NOAA managers believe the amounts would be minimal.

5.4 SS Coast Trader

The freighter Holyoke Bridge was built during the First World War by the Submarine Boat Company at Newark, New Jersey as part of the United States’ response to the war and the need for ships due to losses to U-boats. In the 2 weeks following the United States’ entry into the war in April 1917, German submarines sank 122 ships as part of a calculated strategy of unrestricted warfare. The American response was to declare an emergency and create the U.S. government subsidised Emergency Fleet Corporation (EFC), which the U.S. Shipping Board established in April 1917 to meet the emergency need for more ships (Hurley, 1927). Hundreds of vessels—steel, concrete and wooden-hulled—were laid down and completed during the remaining year of the war in Europe and up into the early 1920s.

Holyoke Bridge, laid down in late 1919, was launched at the end of January 1920. It was a riveted steel-hulled steamship, 324 ft (98.8 m) long, with open cargo holds for bulk commodities or crated goods, and its fuel capacity was 8088 barrels of Bunker C heavy fuel oil. Owned by the U.S. Shipping Board, it was chartered to private companies, all part of an expanded postwar U.S. merchant fleet that worked coastal routes as well as more extended voyages to Central and South America, ports in the Pacific, and transatlantic destinations. During the 1920s and 1930s, renamed SS Point Reyes, and then Coast Trader, the steamer worked out of the Gulf of Mexico and along the U.S. west coast until the outbreak of the Second World War. Chartered to the U.S. Government’s War Shipping Administration, Coast Trader was lost to Japanese submarine attack on June 7, 1942, after it departed Port Angeles, Washington with a cargo of newsprint, bound for San Francisco.

When off the coast of Vancouver Island and past the Straits of Juan de Fuca, Coast Trader was hit with a single torpedo near the stern by the Imperial Japanese Navy submarine I-26; the 56 members of the crew were able to escape the sinking without fatalities as Coast Trader sank by the stern, disappearing 40 min after the explosion. The position of the loss was generally known but not charted, and at the time of the PPW study, Coast Trader was, while never precisely located as a shipwreck, thought to be on the maritime border of the U.S. and Canada and off the coast of Washington’s Olympic Peninsula and British Columbia’s Vancouver Island. While it might have still held a full capacity fuel load of 8088 barrels (339,696 gallons/1,286,889 liters) it was rated a MEDIUM risk:

For the Worst Case Discharge, Coast Trader scores Low with 11 points; for the Most Probable Discharge (10% of the Worse Case volume), Coast Trader also scores Low with 10 points…survivor reports of the sinking make it sound like substantial amounts of oil was lost when the vessel sank, it is not possible to determine with any degree of accuracy what the current condition of the wreck is and how likely the vessel is to contain oil since the shipwreck has never been discovered (NOAA, 2013a:5).

The study also noted that ‘based on the large degree of inaccuracy in the reported sinking location and the depths of water the ship was lost in, it is unlikely that the shipwreck will be intentionally located’ (NOAA, 2013a:5, 34).

At that time, what was not known was that a Canadian Hydrographic Service multibeam sonar survey on October 15, 2010, imaged a shipwreck off the entrance to the Strait of Juan de Fuca, which would prove to be Coast Trader. Jacques Marc of the Underwater Archaeological Society of British Columbia, a regional nonprofit avocational archaeology group that works in close coordination with the Province of British Columbia’s Cultural Branch, assessed the sonar data and concluded the wreck was likely Coast Trader. While this was not known to NOAA at the time of the 2013 report, sharing of data with the Canadian Coast Guard and UASBC led to a decision to deploy an ROV to assess it on a NOAA-funded expedition to the region by the Ocean Exploration Trust vessel E/V Nautilus that would work both in U.S. and Canadian waters in June–July 2016.

E/V Nautilus made a single extended ROV dive on the sonar target on June 2, 2016. A team from NOAA, OET, and U.S. Coast Guard, the Underwater Archaeological Society of British Columbia and the Vancouver Maritime Museum joined via remote telepresence from the University of Rhode Island-based Inner Space Center (Delgado et al., 2018). The wreck lay in 541 ft (165 m) off Victoria, British Columbia. The multi-hour dive verified that the wreck was Coast Trader, which at that time had not been visually inspected and remained officially unidentified.

The wreck was found on its keel, upright on the seabed, oriented on a 346-degree heading. The torpedo detonation tore off the stern and essentially destroyed the freighter from just aft of its midships deckhouse. The bow was intact, and this high degree of structural preservation enabled the team to quickly match features on the wreck to the plans and images of Coast Trader. The forensic inspection of the wreck found that the reported area of the torpedo hit was slightly off; instead of hitting at the area of Hold #4, it hit Hold #3, blowing off the stern, which lay attached by torn steel plating to the intact forward half of the wreck. If fully fueled, Coast Trader lost half of its presumed oil at the time of the torpedo attack. The remaining likely remains inside the well-preserved, lightly corroded forward half.

While the wreck’s lighter steel superstructure and decks are corroded and have failed in places, the hull appears intact and without visible corrosion. The key question is whether Coast Trader will soon collapse and catastrophically release whatever oil remains as opposed to a slower release of oil through vents and piping over time. Based on the visual inspection, the 2016 team did not believe this was either imminent or possible for a number of years. The result of the survey was an archaeological report, but also a revision of the PPW score for Coast Trader. As noted at the time, the dive on Coast Trader added to our understanding of the event provided the means by which a more detailed assessment of the wreck as both a historic site and a potential pollution hazard could be completed. This demonstrated that while wrecks with fuel left inside are a concern, some, like Coast Trader, do not need expensive mitigation for the foreseeable future.

5.5 Conclusion

Desktop assessments have value, especially when a task as daunting as assessing the pollution potential for tens of thousands of shipwrecks is requested. The ultimate selection of 87 vessels as the highest-risk potentially polluting shipwrecks in U.S. waters substantially narrowed the focus, but the work was still considerable. In some cases, insufficient archival information was available, and while some shipwrecks were ‘known,’ insufficient detail of each wreck’s position, damage, structural integrity and change over time was lacking. As noted, detailed assessment would only come with cruises or dives of opportunity. The cruises and investigations ‘of opportunity’ model proved effective, both in assessing risk and also the site characteristics of historic shipwrecks. At the same time, other assessments have found large amounts of oil and remediation was necessary, as in the case of SS Coimbra and SS Munger T. Ball (Brennan et al., 2023), and is most likely with SS Bloody Marsh (Brennan et al., Chap. 9, this volume).

In the three cases presented here, valuable information was obtained. Coast Trader was found to not be in U.S. waters, but in Canada’s. This did not change the assessment in regard to areas of impact, which had primarily been scientifically calculated to be in Canada, but the determination that at least half of the potentially spillable fuel had been discharged in 1942, and the structural integrity of the surviving forward area of the wreck did shift the assessment of risk. Similarly, the realisation that the sinkings and post-loss changes to the wrecks of both Fernstream and Mission San Miguel had also released oil, likely in small amounts over a protracted period, which lessened the risk of a modern major event. The obvious conclusion is that opportunities to assess potentially polluting, historic shipwrecks should be a priority in planning for research cruises. This aspect is now being applied in the United States by NOAA’s Office of Ocean Exploration as a result of the assessments of Fernstream, Mission San Miguel and Coast Trader.