Rather surprisingly for such a hot subject, the Top 100 list of most cited articles in Environmental Toxicology and Chemistry includes only two dealing with the infamous biocide tributyltin (TBT), which was used extensively in marine and freshwater antifouling paints during the final four decades of the twentieth century. The earlier of these articles, by Tolosa et al. 1, is ranked joint ninety-fourth and showed clearly that the highest levels of TBT were associated with environmental samples from docks, marinas, and other concentrations of marine vessels. Furthermore, Tolosa et al.'s observation that significant amounts of TBT contamination were to be found in sediments some years after partial bans had been widely introduced was an ominous hint that problems caused by TBT might be long-lived. We now know that the half-life of TBT in anaerobic sediments can run into decades and that contaminated sediments continue to act as a slow-release source of the substance. The later of these two Top 100 articles is a review of TBT's putative mode(s) of action by Matthiessen and Gibbs 2, and is ranked eighth—a reflection of the fact that the article identified the substance as a prime example of an endocrine disrupting chemical causing demonstrable environmental impacts at a time when such chemicals were becoming big news. The review concluded that aromatase inhibition was a probable cause of TBT-related effects such as imposex in gastropod mollusks, so it is interesting to note that a more likely (or at least additional) explanation is now thought to be an interaction of TBT with a molluskan homologue of the human retinoid X receptor 3.
Stepping back to see the bigger picture, it should be remembered that the use of TBT in antifouling paints for ships started in the early 1960s, and at its height in the 1980s the global market was 2,000 to 3,000 metric tons per annum. Its acute toxicity to mollusks was discovered early on, and it was investigated for use as a tropical molluscicide, but its subtle effects on mollusc reproduction emerged only in the 1980s. In the United Kingdom, penis induction (or imposex) in female dogwhelks (Nucella lapillus) was first reported by Blaber 4 in 1970 and later shown to be caused by TBT that had leached from treated boat hulls or been scoured off in particulate form during repainting 5. In severe cases, imposex in dogwhelks led to local population extinctions 6, and related masculinization and shell-thickening effects in oysters and other mollusk species caused serious problems for the mariculture industry in several countries 7, 8. In the gastropod mollusks alone, it was estimated that about 150 species were adversely affected worldwide 9.
Most of these highly undesirable effects were known by the mid-1980s, and they led to two phases of regulatory action. Initially, it was hoped that restricting the use of TBT to smaller vessels would provide sufficient risk management, on the grounds that such boats are numerically dominant in shallow waters, which provide the least dilution of leachates. For example, a complete ban on the use of TBT on UK vessels of <25 m waterline length was imposed in 1987, and many countries and regions rapidly followed suit. However, continued environmental monitoring showed that TBT from larger vessels was also causing problems (see below), so a global ban affecting all shipping was eventually arranged through the International Maritime Organization. This ban came into effect in 2003 for new applications of TBT paint, with a final deadline of 2008 for complete removal of all TBT from ships' hulls. In other words, the delay between the availability of an essentially complete environmental impact dossier and the final ban was more than 20 years. This was partly due to the unavailability of effective alternatives at the time and partly because it took several years for the International Maritime Organization to obtain ratification of a ban by sufficient countries. Furthermore, an interesting argument deployed by the paint and shipping industries in the 1990s against the withdrawal of TBT was that its use was minimizing release of carbon into the atmosphere by reducing the friction experienced by ships and thereby lessening fuel consumption. This is an instructive lesson about the difficulty of overcoming vested interests in existing chemical products, even when the environmental damage is known to be serious, and shows why environmental risk assessment in advance of commercial use is so important. Of course, TBT entered the market before such assessments became routine.
The question arises as to whether the various regulatory actions led to sensitive organisms recovering from the impacts of TBT. The early national and regional bans on using TBT on small vessels certainly led to improvements in the status of some mollusk species. This is not the place to chronicle these in detail, but the example of dogwhelks N. lapillus will suffice. Birchenough et al. 10 and others monitored recovery in British dogwhelks and showed that recolonization had slowly begun in some areas by the end of the 1990s and that the severity of imposex had widely declined. However, even after the 2003 global ban on new applications, problems with dogwhelks and several other molluscs persisted near centers of shipping activity. For example, TBT residues and imposex could still be detected in Icelandic dogwhelks in 2008, especially near large harbors such as Reykjavik 11. Substantial recovery by 2011 (3 years after the International Maritime Organization deadline for removal of all TBT from ships) of formerly near-extinct dogwhelk populations has, however, recently been reported on the south coast of England between Poole and Selsey 12, and the overall proportion of English coastal sites considered largely free of imposex (Oslo and Paris Commission, assessment classes A and B) has increased from 34.0% in 2004 to 89.4% in 2010 to 2011 (http://chartingprogress.defra.gov.uk), although some problems still remain. Imposex is irreversible; and dogwhelks can live for up to 10 years, so some effects will gradually disappear as older organisms die off. However, a proportion of the continuing effects reported in several species from around the globe—although these effects are generally milder than those seen in the 1980s—is probably due in large part to TBT's essential immortality in anaerobic harbor sediments and other relatively undisturbed marine deposits which act as slow-release sources. In several countries, some harbor sediments are now classified as toxic waste. In these cases, remediation may be prohibitively expensive as dredgings cannot legally be disposed of at sea as in former times.
Perhaps the most interesting issue concerns the degree to which TBT damaged whole communities or ecosystems, because it could be argued that there has been a rather narrow research and monitoring focus on just a few headline species such as dogwhelks and oysters (partly because it has been thought that dogwhelks are the most sensitive organism). Somewhat shockingly, given TBT's former high profile, community-level studies of biodiversity in TBT-contaminated areas have been scarce. The few available published studies, however, have revealed serious community-level impacts that are by no means restricted to mollusks. Using the concept of pollution-induced community tolerance, Blanck and Dahl 13 showed that marine periphyton communities near a Swedish marina had been altered by TBT, although they were significantly recovering in 1991, following a local ban on TBT use on small boats in 1989. The only substantial study of a recovering invertebrate community, however, took place on the River Crouch estuary (UK) in six sampling years between 1987 (when the local small-boat TBT ban came into force) and 2005 inclusive (14, 15 and other papers cited in these publications).
The Crouch was chosen for this large-scale and long-term investigation because almost all TBT inputs to the estuary before 1987 were from small pleasure vessels, so the continuing use on large ships elsewhere could not confound the results. Furthermore, the invertebrate ecology of the Crouch before the TBT era was already documented. The study focused on both the invertebrate benthic infaunal and epifaunal communities and showed that, in parallel with declines in the TBT loads of waters and sediments, the diversity of both subtidal communities increased substantially over 5 years from 1987 to 1992, while smaller changes after that were mainly attributable to natural factors such as shifts in sediment particle size distribution. Recoveries attributed to declining TBT concentrations were particularly marked among the sedentary taxa at the three originally more contaminated upstream stations. For example, the total numbers of epibenthic taxa per trawl at these stations increased from 21 to 30 (mean, 25.2) in 1987 to 1988, to 27 to 44 (mean, 32.8) in 1992 to 1997, an average 30% increase. Only when TBT concentrations in sediment had reached <0.01 to 0.04 mg kg−1 dry weight did these communities appear to approach stability. These recoveries included a resurgence of the native oyster Ostrea edulis population, which had been the basis of a significant local fishery before multiple stressors such as TBT, the very cold winter of 1963, and Bonamia ostreae disease had taken their toll. Furthermore, apparent recoveries from TBT-related impacts were seen in several taxa in addition to mollusks, including crustaceans, polychaete annelids, and ascideans, a result which is perhaps not surprising given that TBT-based paints were originally designed to have a broad-spectrum effect on fouling organisms. It is not known, however, whether all of these recovering species were directly sensitive to TBT or whether some were responding to secondary ecological effects. This 18-year investigation nevertheless suggests that the molluskan focus of much TBT research and monitoring had been too limited. Ironically, that iconic species the dogwhelk had still not recolonized the Crouch by 2005, partly because the habitat is only marginal for this species and partly because N. lapillus has no planktonic dispersal stage—its rate of spread by crawling is about 30 m in an individual's lifetime, and the nearest intact populations are probably several kilometers away.
Many conclusions may be drawn from this story, involving both scientific and regulatory aspects. Perhaps most important is that although some communities are able to make quite rapid recoveries from serious pollution (approximately 5 years in the TBT/Crouch example), even strong regulatory action cannot necessarily provide complete remediation of the damage caused by highly persistent and toxic materials once they have escaped into the environment. The continuing need for rigorous environmental risk assessment of all new chemicals is therefore obvious but nevertheless worth restating for those who have not had firsthand experience of global disasters like that caused by TBT.