Teredo navalis
(naval shipworm)

The bivalve mollusc Teredo navalis has a brownish elongate worm-like body, with the anterior part covered by a small calcareous tube shell acting as a wood-boring instrument. T. navalis has been found worldwide and is accepted as a well-established species in the European, North American and Northwestern Pacific coastal areas. However, its origin is still to be defined and its status is currently accepted as cryptogenic in many areas. The main vectors of species spread are shipping and larval dispersal with currents. The species is both eurythermic and euryhaline, withstanding a wide range of temperatures and salinities. Its life-history strategy of being larviparous (with long larval life) is considered the most effective to survive in patchy ephemeral habitats, such as wood. Due to its boring activity, the species quickly destroys wooden structures submerged in water at depths of 0 to 20 m, posing a great hazard to wooden maritime structures in coastal areas and incurring millions of dollars of damage per year. The recent range expansion of T. navalis into the Baltic Sea suggests species adaptation to lower salinity conditions.

The species is known from all oceans except the Antarctic, but many of those records need verification. The Encyclopaedia of Life (EOL, 2014) indicates the species to be present from -37.6°S to 58.39°N, but teredinids have very patchy distributions depending on the presence of suitable substrata.

In the Northeast Atlantic, the species is found in the Norwegian, North and Baltic Seas and in the coastal Atlantic waters of the UK and France. It is also found in the Mediterranean, Black and Azov Seas. Detailed distribution information is provided by Borges et al. (2014). T. navalis has questionably invaded America from Europe, where it is considered cryptogenic (Carlton, 1992). The species is now found in all coastal waters of North America (Grave, 1928). On the Atlantic coast, it is distributed from Newfoundland southward to Texas and, on the Pacific coast, it is spread from British Columbia (Pendrell Sound) to Southern California (San Francisco Bay) (Carlton, 1992). In the Western Pacific, most records are from the northern area. T. navalis is found along the entire Japanese coast and it has been recorded from the South and East China Seas, Xisha Island, Beibu Gulf, Hainan, Taiwan, Yellow and Bohai Seas. The species is also common in Russian Far East waters (Tsunoda, 1979; Bernard et al., 1993). There are few verified records from the Indian Ocean and South Pacific, but the probability of finding the species there is high. Nair and Saraswathy (1971) indicated the species to be present in the Eastern Indian Ocean and the South Pacific (Indonesia, Burma, New Zealand and Australia), although Sliwa et al. (2009) referred to the Australian distribution of this species as only “recorded through literature surveys”.

A high risk of introduction of T. navalis may arise in areas where suitable hydrological conditions and substrata already exist, or might arise, as a result of climate change. Paalvast and van der Velde (2011b) suggested that global warming may cause dry and warmer summers in Northern Europe, decreasing river discharges, leading to increased salinity upstream in the summer and autumn. This may cause upstream expansion of T. navalis, not only in northwest European estuaries, but in southern Europe as well.

Many authors consider T. navalis a potential invader of the Eastern parts of the Baltic Sea (Iljin, 2010; Borges et al., 2014). According to ecological data modelling (Wreck Protect, 2011), the maximum possible spread of the species until 2020 is along the Swedish coast up to the north to Kalmar strait, and along the German and Polish coasts eastwards up to the Gulf of Gdansk, Poland. Estuaries of large European rivers that flow into the Baltic are under serious threat, because of the low indigenous species richness and intensive international shipping traffic in those waters (Nehring, 2006).

Other European areas at risk of further invasion are the Mediterranean and Black Seas. For example, Iljin (2010) suggests that, after the construction of the Volga-Don shipping canal, the probability of appearance of T. navalis in the Caspian Sea, where some areas are especially favourable to this species (Kudinova-Pasternak, 1957), has increased.

T. navalis inhabits both fixed and floating timber. Any type of untreated wood may be attacked, although hard types, such as oak, can be less affected than, for example, fir wood (Paalvast and van der Velde, 2011a). Tree barks are more resistant, providing longer protection from T. navalis to wooden constructions (Tarasov, 1943).

T. navalis can be found at depths up to 30 m, but is mostly found from the surface to a depth of about 5 m (Tarasov, 1943; Turner, 1966; EOL, 2014). According to Iljin (1992), in Japan, the Azov and Black Seas, T. navalis most often occurs near the bottom, where salinity is higher than at the surface. In the southern part of the Black Sea, the species is found at a depth of 0.5-5 m, on old tree roots and wood pieces on stony-sandy bottoms (Culha, 2010).

Tsunoda (1979) noted that heavier borer infestations are generally found on horizontal surfaces, rather than on vertical ones. Of horizontal surfaces, the upper areas were liable to be attacked more severely than the lower ones.

Genetics

Distel et al. (2011) found that the bivalve subfamilies Teredininae and Bankiinae of the family Teredinidae are non-monophyletic and that the main traits used for their taxonomic diagnosis may be phylogenetically misleading.

Reproductive Biology

Reproductive strategy is very important under patchy, ephemeral habitats, such as wood immersed in water. Multiple species and genera of shipworms, competing under similar conditions, exhibit differing peak recruitment seasons and reproductive modes, with free-spawning, short-term brooding and long-term brooding occurring (MacIntosh et al., 2012). Reproductive mode is a key contributor to recruitment success, with habitat constraints favouring a 'middle of the road' strategy of short duration larval brooding, most effectively balancing fecundity, larval retention and dispersion ability. The reproductive pattern of T. navalis demonstrates a highly effective mode of this strategy (Borges et al., 2014).

T. navalis is sex separate, but adult gender cannot be distinguished externally. Young animals are potentially hermaphroditic and pass through alternating sexual phases during their development (Coe, 1941; Grave, 1942; Nair and Saraswathy, 1971). At temperatures of 20°C, T. navalis becomes sexually mature at 3 weeks old (Culliney, 1975). However, in Japan, the youngest T. navalis found spawning was 6 weeks old and 38 mm in length (Tsunoda, 1979). After fertilization, larvae develop in a brood pouch for 2 weeks before they disperse. As many as 1 to 5 million larvae can brood in the gills of an individual shipworm (Grave, 1928) and several generations can be produced per year. T. navalis produces a planktotrophic larva at an advanced veliger stage, which feeds on plankton during the free-swimming phase (Lebour, 1946). Closer to settlement, the larva forms a shell, which is first single, but soon becomes bivalved (Costello and Henley, 1971). After about 20 days and at a size of approximately 0.25 mm, animals are mature enough to settle on wood. If no suitable substrate is available, they can survive for an additional 2 weeks, allowing larvae to colonize new areas (Scheltema, 1971; Nair and Saraswathy, 1971). After successful settlement, larvae undergo metamorphosis, changing from a juvenile to an adult that will never leave the piece of wood.

Physiology and Phenology

In Pacific waters (Japan), shipworms usually breed for 7 months from June through December, with local settlements in January and February (Tsunoda, 1979). Settlement peak varies with localities and depends on fluctuations of water temperature throughout the year, the amount of wood available for the borers and activity of fouling communities. Overall, in Japan, settlement has been observed all year round (excluding spring months), in water temperatures between 9 and 30°C and salinity between 25.7 and 37 ppt. In Russian waters (Ussuriysky Bay), T. navalis pelagic larvae have been found from June to September in the inner part of the bay, and in August-September in the open part of the bay, where their frequency of occurrence was nearly two times higher than in the inner part (Kulikova et al., 2013). In European waters, in Rotterdam, Netherlands, T. navalis spawns from August until the end of November, with mass infestations taking place from September (Paalvast and van der Velde, 2011a). In Atlantic American waters, the breeding season of T. navalis lasts from May to October, starting at water temperatures of 11-12°C, and spawning occurs several times during the year (Grave, 1928).

Longevity

Life duration of T. navalis ranges from one to three years (Sordyl et al., 1998).

Population Size and Density

Population size and density of T. navalis vary considerably depending on habitat and season. In the Azov Sea, the number of borings per dm² of a saw cut of piles sometimes reached over 100 (Iljin, 2010). In the Japan Sea, near Vladivostok, the number of borings per dm² was initially about 15, while after three years at 2 m depth, it reached 278 per dm² (Iljin, 1992). Near the Swedish coast, the number of borings was 91 per dm² at 0-1 m depth and only 8 per dm² at 1-10 m depth (Norman, 1976). This difference can be explained by the fact that surface waters warm faster than deeper ones. In Northern Germany, the maximum number of borings found was over 10,000 per m² in fir floating fenders, 4600 per m² in oak piles and 200 per m² in fir pier posts (Tuente et al., 2002).

Nutrition

T. navalis has the ability to consume wood as food (xylotrophy) (e.g. Gallager et al., 1981; Distel et al., 2011) and can survive on a wooden diet only (Hoppe, 2002). Symbiotic cellulolytic nitrogen-fixing bacteria (Teredinibacter turnerae) are harboured within specialized epithelial cells (bacteriocytes) located within the gills of T. navalis (Popham and Dikson, 1973; Distel et al., 1991). The unique bacteria produce cellulolytic enzymes that degrade wood to sugar molecules, on which T. navalis feeds, thus eating its way into wood (Distel et al., 2002).

The ability to feed on wood may have arisen just once in Bivalvia, in a single wood-feeding lineage that subsequently diversified into distinct shallow (where T. navalis belongs) and deep water branches, both of which have been broadly successful in colonizing the world’s oceans (Distel et al., 2011).

Wood consumption, however, may not be the main food source for T. navalis, as recent research has argued. Paalvast and van der Velde (2013), who used a stable isotope approach, found that filter feeding on seston may constitute the main nutrition means for the species. The filtration, supporting food, but also an oxygen supply, occurs with the help of the tube-like siphons. By filter feeding, the woodborers receive mainly nutritional lipids and proteins (Mann and Gallager, 1985). However, they can survive and reproduce without plankton feeding for at least two years, because the cellulolytic nitrogen-fixing symbionts also provide an internal source of dissolved nitrogen, incorporating it into essential aminoacids and thus supplementing the host’s protein-deficient diet (Waterbury et al., 1983; Ahuja et al., 2004).

A detailed account of the functional anatomy of the feeding organs and digestion of the species is given by Morton (1970).

Associations

T. navalis colonies are frequently associated with other wood-boring and fouling organisms. For example, wood degradation seems to be slower when Teredo is associated with Limnoria (wood-boring isopods). The settlement of Teredo larvae is inhibited by the “spongy” surface of Limnoria-infested wood (Kofoid and Miller, 1927). Heavy marine fouling also restricts teredinids activity and slows down wood degradation (Weiss, 1948; Nair and Saraswathy, 1971).

An important association with the endosymbiont bacteria Teredinibacter turnerae, located within the gills of T. navalis, allows wood to be degraded into sugar molecules, on which the species feeds (Distel et al., 2002).

Environmental Requirements

Although preferring rather brackish habitats, T. navalis is both eurythermic and euryhaline (Roch, 1932; Nehring, 2006). T. navalis can survive temperatures from 1 to 30°C and salinity from about 7 to nearly 40 ppt (10-35 ppt being optimal), thus outcompeting all European teredinids (Borges et al., 2014). However, the upper and lower limits of the suggested salinity range may significantly affect T. navalis life cycle, thus restricting its distribution.

In laboratory experiments, adult T. navalis survive at salinities as low as 2 ppt for 24 days, although their activity decreases abruptly below 9 ppt (Blum, 1922). Kudinova-Pasternak (1958) found that adult T. navalis can survive inside wood in freshwater for 3 weeks. If water temperature was close to the animal’s lower or upper range limit, the mollusc died on the 8th or 3rd day, respectively. An increase in water flow also reduced the survival period of T. navalis to 12-14 days. In a population in Sweden, Spicer and Strömberg (2003) described that T. navalis withdraw their siphons and seal their burrows at salinities from 4 to 8 ppt. At salinities less than 4 ppt, burrows can remain closed for at least 6 days, at 22°C, with siphons re-emerging within minutes of a return to 4 ppt. This may be explained by metabolic responses to low salinity: the oxygen uptake of excised gill tissue decreased with decreasing salinity. For larvae, lethal salinity in Bremerhaven harbours, northern Germany, was 5 ppt (Tuente et al., 2002).

The lower limit of water temperature that T. navalis adults can withstand varies considerably according to different publications (Roch, 1940; Tarasov, 1943; Ryabchikov and Nikolaeva, 1963; Ryabchikov et al., 1963; Kudinova-Pasternak, 1971). For T. navalis larvae, temperatures lower than 10°С and higher than 30°C are lethal. The lowest water temperature that allows larvae settling is 15-17°C (Iljin, 2010). Boring activity of the mollusc decreases at 10°C and stops at 5°C (Norman, 1977a, b).

Т. navalis can survive more than 25-27 days inside wood that is not immersed in water, depending on air temperature and humidity conditions (Lane, 1959; Iljin, 1992).

Ice cover can create favourable thermal conditions for the woodborer, because it prevents bottom water layers from fast cooling, leading to the formation of temperature and salinity conditions more suitable for these animals. This is believed to be the cause of high abundance of borer molluscs, and T. navalis in particular, in areas with complete ice cover, and their absence or low abundance in areas without such cover (Ryabchikov, 1957).

Many protozoa (Nair and Saraswathy, 1971), fungi (Kohlmeyer, 1969) and Haplosporidiidae (McGovern and Burreson, 1989; 1990; Hillman et al., 1990) have been associated with teredinids, although their parasitic role remains under studied. The parasitic Atlantic Ciliophora Boveria teredinidi has been recorded from the gills (ctenidia) of T. navalis in San Francisco Bay (Pickard, 1927). Polychaetes are also often found in association with Teredo (Hoagland and Turner, 1980), with some species predating on teredinids (Roch, 1940; Tarasov, 1943).

The detection of new invasive teredinid species in coastal waters is of particular importance. However, they are hard to detect both in the early and late colonization phases, because living inside wood makes them inconspicuous to most types of surveys.

Shipworms enter the wood as larvae, producing minute holes in the wood surface (about 1 mm), hardly visible to the naked eye (Hoppe, 2002). Once inside the wood, the complex tunnels they excavate are not visible from the exterior, meaning the organisms can go undetected until the wood is heavily infected (Turner, 1966).

Tunnels inside the wood are up to 1 cm in diameter and can be found just below the surface of the infected wood. These tunnels are numerous and give the wood a characteristic lace-like appearance and a sponge-like structure. Inspection of tunnelled wood and of animals under the microscope allows infestations of teredinids to be detected and the species to be identified. Identification guides, such as those by Turner (1966; 1971), can be very useful.

The mollusc can be found with the aid of divers, who look for siphons protruding from wooden structures (Nair and Saraswathy, 1971). Stethoscopes and microphones can also be used to detect the rasping sound (boring activity) of the teredinids.

Several methods have been suggested to estimate the degree of species infestation: to saw the infected wood every 1 m and count the number and diameter of holes (Iljin, 1992); to count the number of surface entrance holes (Dons, 1940); to estimate weight loss of the infected wood (Iljin, 1992); to measure wood resistance (Bozich, 1939); and others. More recently, Sivrikaya et al. (2012) found that using a non-destructive measurement of dynamic modulus of elasticity (MOE) is very effective for evaluation of the degree of wooden material deterioration. The MOE showed a very good correlation with the deterioration determined by visual assessment both in untreated and treated wood. Dynamic MOE allows rapid on-site evaluation of infestation, rather than requiring measurements in a laboratory and without causing damage to the structures being evaluated.

It is important to distinguish boring activities of teredinids from those of some crustaceans, such as species from the genus Limnoria (Isopoda). Limnorid tunnels are smaller, usually interconnecting, and have a series of pinhole-sized punctures along their length (CABI, 2018).

The Teredo genus includes about 30 species that live in very particular conditions, i.e. inside wooden material, such as logs, pilings and ships. Because of this, the woodborers form a particular morphotype that is very similar in different species, making teredinids a difficult group to identify from morphological features. Identification is based almost entirely on the morphology of the pallets, shell details and siphons morphology. All species of the genus retain the young to veliger stage and have largely calcareous pallets, variable in shape, but with the blade always in one piece, usually with a small cup that may be divided medially (Turner, 1966). Turner (1966) provides identification keys to teredinids, where different species are illustrated. Rowley (2005) suggests to use the following characteristics for identification of T. navalis in the field: small white shell; light brown soft body tissues; reduced trilobed shell up to 2 cm in length; tube up to 60 cm long and 0.8 cm in diameter; anterior and posterior lobes of the shells similar in size; shells triangular in shape.

A combined use of morphology, DNA barcoding and nuclear locus sequences has been shown to be useful in investigating the taxonomy and systematics of Teredinidae (Borges et al., 2012).

Due to the variable regulations around (de)registration of pesticides, your national list of registered pesticides or relevant authority should be consulted to determine which products are legally allowed for use in your country when considering chemical control. Pesticides should always be used in a lawful manner, consistent with the product's label.

Prevention

Historically, no completely effective prevention methods against T. navalis have been identified. In ancient times, the hulls of wooden ships used to be protected from woodborers with copper sheathing and, before that, with tar. In the Netherlands, after a large dike collapse in 1731 due to woodborers activity, the only solution was a major change in the dike construction, which was rebuilt from imported stone (Hubschman, 1979). Despite obvious progress in the study of T. navalis, no effective protection measures have so far been developed.

Chemical defence has been widely used. Historically, creosote (composed mainly of phenol derivatives) is a popular chemical means of defence against T. navalis. Poles impregnated with creosote can resist 10 years in the Caribbean Sea, up to 20 years near California, up to 15-30 years near San Francisco, and even up to 50 years in other USA waters (Iljin, 1992). In the Black Sea, near Novorossiysk, treated poles remained intact for 13 years (Ryabchikov, 1957).

Chemical impregnation with copper chemicals, such as chrome copper arsenate (CCA) or borax (CKB) may also temporarily prevent wood from deterioration. Wood treated with copper compounds can be defended from teredinids for 15 or even up to 40 years. Hormonal treatment can affect some stages of the bivalve’s life cycle, such as settlement, metamorphosis and shell formation, preventing normal development and, therefore, heavy infestation (Turner, 1976).

Humar and Lesar (2013) investigated performance of copper-ethanolamine-treated wood exposed to sea water at the Port of Koper, Slovenia. The authors found that, while untreated wood was completely degraded after 10 months of exposure, chemical impregnation prolonged the service life of wood. Even using the lowest concentration of preservative solution (0.31%), wood samples were only slightly decayed, whereas wood impregnated with higher copper concentrations (> 0.31%) showed almost no defects after exposure to marine borers.

In the Black Sea, untreated wooden samples, particularly of Scots pine, were severely attacked by T. navalis, while samples of oak and chestnut treated by copper-chromium-arsenic (CCA) were moderately attacked and treated samples of Scots pine sapwood and heartwood were sound (Sivrikaya et al., 2012).

Alongside chemicals, some simple non-chemical defence methods may be surprisingly effective. For example, freshwater treatment, by keeping wooden ships in freshwater rivers, estuaries or lagoons for several months; airing for a month, or keeping the ship in quarantine; utilization or elimination of sunken wood in ports; functional construction engineering; using temperatures that are lethal to woodborers; use of naturally resilient types of wood, such as barked wood. Tsunoda (1979) considered landing of wooden material as one of the most reliable ways to protect logs against shipworm attack. Covering in a thin layer of metal is also a widely used protective method. More recent research has involved the use of a fiberglass and polymer composite shield to protect wooden piers and marine structures (Elam, 2009).

Valuable shipwrecks can be protected by being wrapped in different types of textiles, thus providing a physical barrier to prevent access by the organisms, without totally preventing water movement. This method has proven to be cheap, easily applied and to have a low environmental impact. Eriksen et al. (2014) examined the ability of two plastic materials (geotextile fabric and a plastic membrane) to stop both initial attack by T. navalis and their effects on archaeological wood. Blocks of pinewood were submerged in the southern part of the Kattegat, Denmark. After settling and attack by T. navalis had been confirmed, the blocks were removed and wrapped in either a geotextile fabric (polypropylene and polyethylene) or in a proprietary plastic membrane (polyethylene). After one week, oxygen levels around the test blocks wrapped in plastic membrane had dropped drastically and led to the death of all shipworms. Although no new shipworms attacked the wood wrapped in geotextile fabric, it did not prevent the passage of oxygenated seawater, as living individuals were found in the test blocks after 46 weeks of wrapping and submersion.

The only efficient way to prevent economical losses caused by Teredo is to build coastal protection and submerged constructions using non-wood materials, such as stone, concrete or plastic.

Control

Physical/Mechanical Control

Using electricity has been suggested for elimination of woodborers (Iljin, 1992). Norman and Henningsson (1975) described positive results of using explosives in elimination of wood-boring organisms: non-treated poles were destroyed in 3 years, while poles in the area where explosives were used remained intact for 5 years.

02/06/14 Original text by:

Ekaterina Shalaeva, Consultant, London, UK