1.1 Overview

Shipwrecks lie hidden below the surface of the water and, especially those in deep water, are out of sight and easily ignored. Many of these wrecks from the modern era either contain or are suspected to contain hazardous materials that are within the metal hulls which have the potential to cause an environmental disaster should they leak or spill. A potentially polluting wreck (PPW) is a shipwreck containing a cargo or a large volume of its own fuel that remains within the wreck and has the potential to cause an environmental hazard should the structure become compromised and either leak or catastrophically release. This book addresses those wrecks with the potential to pollute due to petroleum cargoes or bunkers. While unexploded ordnance and munitions also represent both a hazard and toxic substance, addressing this type of cargo and threat is different than for petroleum cargos and would need its own volume on the subject. According to the International Union for the Conservation of Nature (IUCN), ‘marine pollution from sunken vessels is predicted to reach its highest level this decade, with over 8,500 shipwrecks at risk of leaking’ (IUCN, 2023). They contain hazardous materials including chemicals, unexploded ordinances, and an estimated six billion gallons of heavy fuel oil. ‘This is 545 times more oil than the Exxon Valdez leak in 1989 and 30 times that of the Deepwater Horizon spill in 2010, both of which had severe and long-lasting environmental consequences’ (IUCN, 2023). Many of the wrecks identified as PPWs are those sunk during the two world wars, particularly oil tankers, but also include freighters, as well as ships from parts of the twentieth century that foundered in storms.

A study in 2005 for the International Oil Spill Conference compiled a worldwide dataset of PPWs and identified 8569 potentially polluting wrecks with the parameters of tankers greater than 150 gross tons (GT) (1583) and non-tank vessels greater than 400 GT (6986) (Michel et al., 2005: 69). The potential environmental risks these wrecks pose are significant. The common example of SS Jacob Luckenbach, sunk in a collision off San Francisco’s Golden Gate in 1953, was estimated to have killed thousands of birds over the course of a decade of mystery oil slicks from the wreck with untold other impacts to the marine environment. Only in the aftermath of the Deepwater Horizon spill and the research conducted in the Gulf of Mexico since do we have a better understanding of some of the environmental impacts of such disasters to the deep-sea ecosystem.

The main question regarding potentially polluting wrecks is, of course, what is the potential for a major spill? This subject has been written about extensively and excellent discussions are presented by Michel et al. (2005) and Landquist et al. (2013). While there is no clear consensus on the best comprehensive approach for risk remediation between coastal states who have dealt with the issue, a collective approach to the problem is incident specific and requires evaluation of the wreck site, environmental conditions, methods of risk reduction, and potential for catastrophic pollutant release versus slow leaks (Landquist et al., 2013: 91; Etkin et al., 2009). This is particularly true for older historic shipwrecks. ‘It is clear that most of the oil remaining on these wrecks will eventually be released. More than 75 percent of the wrecks date back to World War II and thus have been underwater for 55–65 years, so there is added concern that corrosion will lead to increased oil discharges’ (Michel et al., 2005: iv).

The sheer number of losses of wrecks that both polluted at the time of their loss and those which retain the potential to pollute is extensive. Germany’s U-boat campaign, the ‘Battle of the Atlantic,’ sank some 3500 merchant ships, 175 Allied warships, and lost 765 U-boats in European, North Atlantic, Gulf of Mexico and Caribbean waters (Delgado, 2019: 337–338). Total allied warship losses in World War II, globally, neared 2000 vessels. In the Pacific and Indian Oceans, allied forces sank 686 Japanese warships, and 2346 merchant vessels during World War II (Joint Army-Navy Assessment Committee, 1947). As the corrosion of thousands of these steel hulls in the marine environment persists, the risk of accelerated leaking or catastrophic release increases with time in nearly every ocean on the planet as a result of that global conflict.

The determination of a shipwreck as environmentally hazardous or potentially polluting does not, however, negate its significance as a historic site. While the pollution risk from PPWs can pose a threat to ocean heritage in the environmental sense, the potential hazard also poses a risk to the archaeological integrity of the historic sites. NOAA’s Remediation of Underwater Legacy Environmental Threats (RULET) project was designed to assist ‘in prioritizing potential threats to ecological and socio-economic resources while at the same time assessing the historical and cultural significance of these nonrenewable cultural resources’ for shipwrecks in US waters (NOAA, 2013: 1). Consistent with the National Historic Preservation Act, RULET not only assessed which shipwrecks in US waters might pose a risk of polluting, but also addressed whether they met the standard for heritage listing.

Thankfully, not only for heritage reasons but also because some of these wrecks contain human remains, they are rightly seen as war grave sites, calling for respectful treatment even under challenging circumstances. The USS Arizona provides a good example of one approach, a memorial park in which there has been no intervention to remove pollutants, but the National Park Service periodically cleans up the sheen (see Glover, Chap. 4, this volume). Another example is the UK decision to remove the oil from the HMS Royal Oak after an environmental assessment and long deliberations on the desire not to disturb the war grave, unless absolutely necessary to address the threat to the marine environment, and the livelihoods dependent upon a clean ocean (see Hill et al., Chap. 6, this volume).

Similarly, at the site of USS Mississinewa at Ulithi Atoll, typical approaches to oil removal have involved in depth assessments of the wreck sites’ condition, stability, and integrity of the hulls, which enables archaeological documentation and characterisation. Methods of tapping hulls at locations over cargo tanks and pumping petroleum cargoes do not usually compromise the stability of the structure and rarely impact the sites’ archaeological integrity or disturb any human remains should they exist within the hull. In determining how to remediate pollution risks, as each site is a case-by-case matter dependent on environmental conditions and site parameters, however, the potential environmental hazard must avoid or minimize harm to the integrity of the historic site. The in situ preservation policy that prefers non-intrusive activities is in harmony with the respectful treatment of these wrecks that are also war graves.

While a shipwreck site comes into equilibrium over time with the marine environment, corrosion is a constant process and steel hulled ships will eventually rust away; at some point throughout the lifecycle of an oil tanker shipwreck, there will be a release of its cargo. This is an example of an extracting filter, which removed elements of a ship from the site during its formation process, as defined by Muckelroy (1978). Bottom fishing gear has been shown to severely impact wrecks and has likely caused oil release events due to strikes (e.g., Brennan, 2016; Delgado et al., 2018; Brennan et al., 2023), and is a more event-based extracting filter compared to corrosion, which is constant. While shipwrecks in deep water are out of reach of such anthropogenic impacts such as trawl and dredge strikes, corrosion remains a factor that will in time cause petroleum cargoes to release. And, with the advent of deep-sea mining, even PPWs in areas under the high seas are at risk of an inadvertent spill, harm to the historic wreck and possible disturbance of a war grave.

The imperative factor in assessing and mitigating pollution risk from PPW sites, and in characterising their historic and archaeological integrity, is accessing them, a process that begins with locating unfound wrecks and being able to get divers or ROVs on site to systematically document them. Larger scale ocean mapping and exploration is needed to find many expected PPWs that lie in deep water. This can then be followed by full assessment and characterisation of the sites as an initial baseline from which further corrosion and hull integrity can be assessed in the future. In the modern era, we have better technology for surveying and accessing the deep sea, detecting shipwreck sites on the seabed, and post-Deepwater Horizon, better modeling for how oil behaves in the marine environment following a spill. The combination of modern technology and the time that has passed since World War II that left these wrecks on the seabed poses an increasing pollution risk that we have the means to address, if resources were devoted to proactive mitigation.

1.2 RUST, RULET, and the Potentially Polluting Wrecks Study

The Potentially Polluting Wrecks effort at NOAA for shipwrecks in US waters began with the compilation of a database known as Resources and UnderSea Threats (RUST) in 2002 by the Office of National Marine Sanctuaries following a series of mystery oil spills in sanctuaries waters, and which led to the identification of oil from the wreck of Jacob Luckenbach on California beaches. The RUST database was designed to be incorporated into the Sanctuaries Hazards Incident Emergency Logistics Database System (SHIELDS) developed by NOAA’s Office of Response and Restoration’s Hazardous Materials Division. ‘The mission of the RUST database is to develop, maintain, and manage an active and comprehensive inventory of undersea threats and potential environmental hazards within United States waters’ (Overfield, 2004: 74). Following its establishment within the Office of National Marine Sanctuaries, RUST was expanded to include ‘all marine waters in US coastal zone and EEZ’ (Zelo et al., 2005: 807). This work built upon previous PPW work from the International Oil Spill Conference as reported by Michel et al. (2005).

A congressional appropriation in 2010 authorised the RUST database to be expanded and evaluated to identify the most significant environmental threats from shipwrecks in US waters. This led to the Remediation of Underwater Legacy Environmental Threats (RULET) project, which correlated physical, historical, and environmental data with the shipwrecks in RUST and analyzed those data to determine which wrecks posed pollution risks. As RUST was ‘not an effective risk assessment tool in its original form [because] it did not address the larger question of how to characterise potential pollution threats,’ (Symons et al., 2014: 784) additional data and analysis was needed. The RULET project narrowed down some 20,000 wrecks to 573 within the US EEZ that could pose a substantial pollution threat, and then further to 87 ‘potentially polluting wrecks’ or PPWs based on modeling of potential oil release based on worst case and probable discharge (NOAA, 2013; Symons et al., 2014). The final report in 2013 was the Risk Assessment for Potentially Polluting Wrecks in U.S. Waters (NOAA, 2013) and was accompanied by a Screening Level Risk Assessment Package for each of the 87 shipwrecks. This project was conducted within NOAA’s Office of National Marine Sanctuaries, working with the Office of Restoration and Response, NOAA’s Maritime Heritage Program, and the U.S. Coast Guard.

In order to refine the list of shipwrecks in US waters that could pose an environmental risk, the RULET study used a series of risk factors and data to assess the potential structural integrity of the wrecks and the likely amounts of oil remaining on board, and factors that might influence removal operations. These include the amount of oil on board including both cargo and bunker fuel, if the wreck may have been demolished as a risk to navigation, if a large amount of oil was likely lost during the sinking, and the nature of the sinking event (NOAA, 2013: ES-3). Response recommendations, in turn, included a variety of options from leaving shipwrecks alone and responding to leaks or spills ad hoc, to full oil removal and salvage. Intermediate options include background research to develop wreck-specific response plans, site monitoring, and in situ assessments through remote sensing and ROV observation (NOAA, 2013: 74).

Factors affecting planning response options include the type and volume of oil on board, water depth and visibility, location relative to shore, and other site factors like the wreck orientation and condition. Importantly, the study addresses the potential impact of remediation on shipwrecks’ historical integrity: ‘this vessel is of historic significance and will require appropriate actions under the National Historic Preservation Act (NHPA) and the Sunken Military Craft Act (SMCA) prior to any actions that could impact the integrity of the vessel’ (NOAA, 2013: 6). Such assessments are essential to protecting shipwrecks that are historically significant and war graves. However, in many of the cases in the NOAA PPW study, the wrecks had not yet been located, so the response options were left high-level and all encompassing.

Following the release of the study, three subsequent actions by NOAA further refined the study. The first was defining a process by which a legal determination of historical and archaeological significance, through the existing program of the US Government, the National Register of Historic Places. The second was, through opportunistic studies by NOAA and partners, conducting physical, non-disturbance assessment of some of the 87 high risk wrecks. The first was the freighter SS Fernstream, sunk in a collision in 1952 near San Francisco Bay’s Golden Gate Bridge; the second was the SS Coast Trader, sunk off the coast of British Columbia in 1942. The third was the USNS Mission San Miguel, a World War II era naval tanker lost at Maro Reef in the Northwestern Hawaiian Islands in 1957 (see Delgado, Chap. 5, this volume. As a result of those assessments, the potential for risk was downgraded for all three wrecks. Concurrently, NOAA’s Office of National Marine Sanctuaries, working with partners, was also conducting a detailed survey and assessment of Battle of the Atlantic losses off the coast of North Carolina as part of a proposed expansion of USS Monitor National Marine Sanctuary between 2008 and 2014. These detailed assessments of wrecks, some of them listed in the PPW study but not determined to be high-risk, nonetheless were key in developing the right level of documentation for World War II PPW sites for heritage assessment, especially four German U-boats, two Allied warships, and several merchant ship losses in shallow and deep water (Delgado, 2019: 339–342). The RUST/RULET and Potentially Polluting Wrecks databases and studies produced an important assessment of the wrecks in US waters and helped compile information on the wrecks, especially those yet to be found, as a vital tool in beginning the efforts to locate them.

1.3 Site Assessment and Pollution Risk Remediation

The waging of two world wars over the course of three decades of the early twentieth century left oil-filled hulls on the bottom of the sea across the entire globe. Aside from those in US waters, high concentrations of potentially polluting wrecks are in the waters surrounding the United Kingdom, the Baltic Sea, and a large swath of the South Pacific. As these hulls corrode and begin to leak, coastal states have had to determine how best to remediate and mitigate these pollution risks. The International Oil Spill Conference study reflected on the growing demand for proactive oil removal from wrecks containing petroleum cargoes rather than the common reactive approach to address the environmental hazards once oil had begun to leak (Michel et al., 2005: 1). Like NOAA’s approach, Michel and colleagues addressed the multifaceted approach required to assess and mitigate oil spill risk from wrecks, including spill modeling, costs and technology for diver versus ROV work, vessel needs, oil disposal, and the mechanics of removing heavy fuel oil in colder deep water. This global assessment also illustrated the significant spike in potentially polluting wrecks sunk during World War II, defined as those sunk between 60–70 years ago, in steel hulls corroding ever since in the ocean (Michel et al., 2005: 17); and nearly 20 years have passed since this study. As this study concludes, ‘It is clear that most of the oil remaining on these wrecks will eventually be released’ (Michel et al., 2005: iv).

Expanding efforts and funding for ocean exploration to locate and assess potentially polluting wrecks is an overarching need to investigate the myriad factors that impact a shipwreck site’s condition, ability to retain oil cargo, and ultimately release it. Liddell and Skelhorn (2018: 83) list the important factors to assess as the type of wreck (i.e., warship or merchant vessel), the nature of the sinking event (the number of torpedo strikes, for instance, or how intact might the hull be), natural factors such as corrosion, and anthropogenic factors such as trawling and salvage. The UK government and Ministry of Defence (MOD) have dealt with the subject of polluting wrecks extensively due to having one of the largest inventories of state-owned wrecks in the world due to its large maritime role in both world wars, as well as the UK’s policy of non-abandonment (Liddell & Skelhorn, 2018: 84). Leaking oil from the World War I battleship wreck of HMS Royal Oak in Scapa Flow led the UK to move from a response policy to one of proactive risk management and the management of the UK inventory of wrecks by Salvage and Marine Operations (SALMO) starting in 2008. This approach of assessing wreck sites and mitigating pollution risk proactively has also recently been done by the United States, as previously noted, with the remediation of the wrecks of USS Mississinewa at Ulithi Atoll in 2003, USS Chehalis in Pago Pago harbor, American Samoa, the German cruiser Prinz Eugen at Kwajalein (Naval Sea Systems Command, 2003, 2011, 2019), and the oil tankers Coimbra and Munger T. Ball off Long Island and Key West (Brennan et al., 2023).

The proactive approach to remediating chronically leaking oil tanker wrecks was recently conducted by the US Coast Guard and Resolve Marine Group on two shipwrecks on NOAA’s PPW list, Coimbra and Munger T. Ball. Both wrecks were known to have been leaking for decades and therefore high priority targets for remediation. Tapping of the oil tanks through the mostly capsized hulls by divers succeeded in removing much of the petroleum product on board the wrecks. Also important to the efficacy of on-site wreck assessments was that these two projects had the author on site for the ROV and diver assessments, which was the first such projects in U.S. waters to have a maritime archaeologist as part of the on-board team. This work also succeeded in correcting a misidentification of the Munger T. Ball wreck as that of tanker Joseph M. Cudahy, which was sunk on the same day in 1942 by U-507, but which had different cargo capacities and carried different fuel types (Brennan et al., 2023).

Prioritisation of PPW remediation operations is another difficult and complex task. Understanding the potential hazards and prioritising which wrecks pose the greatest risk is paramount, as ‘it is economically unfeasible and impractical to remediate all sunken shipwrecks’ (Landquist et al., 2013: 86). A study by members of the British MOD and colleagues called for better standardisation of pollution risk assessment ‘methodologies at both national and international levels’ to develop more ‘reliable means of prioritizing PPW’ (Goodsir et al., 2019: 291). Such standard approaches could help in meeting goals set forth by a variety of conventions that call for protection of both Underwater Cultural Heritage (UCH) sites and marine environments, such as the UNESCO convention of 2001 and the Nairobi Convention of 2007, the latter of which additionally addresses the legal obligation for the removal of pollution hazards from the marine environment (Goodsir et al., 2019; see Aznar and Varmer, Chap. 2, this volume). The staged approach put forth for PPW site assessments is similar to that developed by NOAA and includes environmental desk-based assessment (E-DBA) followed by on-site wreck integrity evaluation along with environmental characterisation, followed by intervention and remediation as the final stage when determined necessary (Goodsir et al., 2019). Proposed responses included to both chronic and acute leak scenarios taking place over continual or catastrophic leak events and acknowledge the political pressures that are often applied to oil spill incidents, but which can be avoided through such proactive approaches.

The work presented by Carter and colleagues follows on previous prioritisation work for PPWs and presents an approach for the span of World War II wrecks in the Pacific. While the removal of petroleum product from PPW sites is sometimes undertaken, the initial desktop assessment and prioritisation is conducted irrespective of that remediation action; the desktop and then on-site assessments are used to assist in future remediations, but on the first level, ‘help prioritize efforts towards more effective management and allocation of resources’ (Carter et al., 2021: 7). This was conducted on the numerous World War II wrecks sunk at Chuuk lagoon, which, ahead of any remediation efforts, have identified 17 of the wrecks as the highest risk, for example, Rio de Janeiro Maru, a Japanese auxiliary ship sunk during Operation Hailstone in 1944. On-site assessments can also take into account factors in addition to cargo, vessel type, and depth, such as oceanographic parameters and the levels of corrosion present on the hull (Carter et al., 2021). Corrosion in the marine environment is the main driver of accelerated pollution risk as the steel hulls containing fuel and liquid cargoes deteriorate (e.g., Macleod, 2002, Glover, Chap. 4, this volume). Diver and ROV inspection of the hull of Coimbra in 2019, for example, found that the leaks were primarily coming from corroded rivets (Brennan et al., 2023).

1.4 Protecting Our Ocean Heritage

Finding, assessing, and potentially mitigating the shipwreck sites of potentially polluting wrecks is an imperative step toward managing and protecting our ocean heritage. The subject of Potentially Polluting Wrecks, and the topic of this volume, is central to the need to protect ocean heritage; PPWs are both historic sites of cultural significance and pose a threat to the marine environment, a juxtaposition that complicates protecting both cultural resource and marine ecosystems. Ocean heritage is both cultural and natural heritage in the material culture from past human activity and humanity’s use of the maritime landscapes throughout history to the present day and into the future. Addressing the risk of pollution events into the marine ecosystem from PPWs works toward protecting the marine environment from further anthropogenic spill events while also documenting and assessing the wrecks as historic sites and, in many cases, war graves. As previously stated, the imperative step toward addressing PPWs is increased ocean exploration and mapping to locate undiscovered sunken ships that may pose a pollution risk so that steps toward assessing and potentially mitigating those risks can be undertaken, along with archaeological documentation of the wrecks for preservation and chronicling the wrecks’ histories.

The chapters that follow will cover the legal context, environmental impacts, archaeology, and case studies of PPWs in certain areas of the Atlantic, Pacific, and the Baltic Sea. The book does not have case studies in all the areas where PPWs exist, and future work to address pollution risks will expand the geographic scope of the subject. Please contact The Ocean Foundation if you are interested in doing education and outreach on PPWs.