The human-mediated spread of marine species from their natural ranges to new habitats is occurring with increasing frequency (Seebens, Gastner & Blasius 2013) and resulting invasions have been reported from most regions (Molnar et al. 2008). These invasions result primarily from shipping, the main vector being ship hull fouling (Hewitt, Gollasch & Minchin 2009; Mead et al. 2011a). Other shipping-related vectors include ballast water (Klein et al. 2010) and sediment retained in ballast tanks (Briski, Bailey & MacIsaac 2011). The increase in the number of ships and ballast tank size (Carlton 1996), accompanied by an increase in transoceanic travel (Ruiz et al. 1997) over the last century, has led to an increase in the introduction and rate of establishment of invasive marine species around the world (Wonham et al. 2000). This is cause for concern as these alien species can pose a threat to indigenous biota (Gurevitch & Padilla 2004; Molnar et al. 2008; Strain & Johnson 2013) and can have negative economic impacts (Cinar 2013; Pimentel, Zuniga & Morrison 2005).

In South Africa, 90 introduced and 39 cryptogenic species have been recognised to date (Bolton, Andreakis & Anderson 2011; Clark & Griffiths 2012; Mead et al. 2011b; Peters et al. 2014), although this is likely to be an underestimation as alien species are continuously being introduced into the country. Most introductions are recorded along the cool-temperate west coast (Mead et al. 2011a); most occurring within harbours, a moderate number on rocky shores and in estuaries and very few offshore (Mead et al. 2011a). Following global trends, most species alien to South African waters have been introduced via hull fouling and ballast water (Mead et al. 2013).

The West Coast National Park (WCNP) is situated in the Saldanha Bay system on the cold west coast of South Africa and includes Langebaan Lagoon (Robinson, Griffiths & Kruger 2004). It was established in 1985 and is the only national park for the protection and conservation of marine species north of the Cape Peninsula (Robinson et al. 2004). The WCNP faces numerous threats, including increased pollution within the bay owing to industrial development and urbanisation (Kruger et al. 2005), expanding mariculture (Olivier, Heinecken & Jackson 2013) and the introduction of marine alien species associated with the port (Robinson et al. 2004). Presently more alien marine species occur within the Saldanha Bay–Langebaan Lagoon system than in any other location along the South African coast (Mead et al. 2011a).

The alien anemone Sagartia ornata was first recorded along Langebaan Lagoon's western shoreline in 2001 in the upper intertidal zone (Acuña, Excoffon & Griffiths 2004). Here, S. ornata was found intertidally on sandy shores partially covering the underlying rocks and in Spartina maritima beds (Robinson et al. 2004), where its densities reached up to 426 ± 81 (s.d.) individuals per m² (Robinson et al. 2005). Sagartia ornata's introduction to the lagoon is thought to have occurred unintentionally through shipping via the harbour of Saldanha Bay (Robinson et al. 2004). It is most likely that the introduction is recent, as extensive annual sampling of the lagoon failed to detect this anemone until 2001 (Acuña et al. 2004). This species originates from Western Europe, the Mediterranean and Britain (Robinson et al. 2004). There it is found widely occurring on kelp holdfasts and in crevices on rocky shores (Gibson, Hextall & Rogers 2001). Thus, the South African west coast, known for its cold water, rocky shores and extensive kelp beds, is very similar to S. ornata's native home range (Acuña et al. 2004) and much potential exists for this species to spread along this coast. Despite this, S. ornata's distribution and abundance was last assessed more than 10 years ago (Robinson et al. 2004) and its impact on indigenous sandy-shore communities has not been considered. This is a notable gap in current knowledge as sea anemones can play an important role in marine communities through their role as predators (Kruger & Griffiths 1998).

Against this background, this study aimed firstly to determine the current status and distribution of S. ornata, secondly to assess the diet of this anemone so as to establish its role as an alien predator and thirdly to investigate its impact on the structure of the sandy-shore communities of invaded areas in Langebaan Lagoon.

Sagartia ornata has a wide native distribution that spans the Mediterranean, Britain and Western Europe (Acuña et al. 2004). This species was first recorded in South Africa in 2001, at which time it occurred only along Langebaan Lagoon's western shoreline within the WCNP (Robinson et al. 2004). Following this initial work, no follow-up studies have been conducted until now. This study found that S. ornata was still restricted to Langebaan Lagoon, although its distribution within the lagoon had changed.

Despite S. ornata living on rocky shores and in kelp holdfasts in its native habitat (Gibson et al. 2001), it occurs in different habitats in South Africa. In Langebaan Lagoon, S. ornata has been recorded in shallow sand partially covering underlying rocks, in S. maritima beds (Robinson et al. 2004), attached to loose rocks on fossilised oyster beds and in beds of the Cape eelgrass N. capensis. This shift in habitat is not uncommon for invading species. For example, the invasive zebra mussel Dreissena polymorpha that occurs on hard substrates in rivers and lakes in its native range occurs on soft sediments in its invaded range (Berkman et al. 2000). This exploitation of new habitats might reflect the adaptive ability often expressed by alien species (Berkman et al. 2000). The spread of S. ornata to new areas within the park since 2001 is notable and is linked to its exploitation of new habitats (i.e. in N. capensis beds and the rather cryptic habitat of loose rocks resting on fossilised oyster beds). It is also possible that S. ornata might have been overlooked in the latter habitat in the 2001 survey, as it is not the type of habitat that was anticipated to support this anemone. It is unlikely, however, that this anemone will spread extensively within the lagoon because of the relatively restricted extent of the habitats in which it has been recorded. In addition to changes in distribution, the abundance of S. ornata has also changed, as more anemones were recorded during 2013 than during the 2001 survey. However, the reason for this change is not obvious. Whilst the 2001 and 2013 surveys were conducted at the same time of the year (i.e. seasonal variation is unlikely to be of importance), routine monitoring of this anemone may reveal the drivers of the observed changes in abundance.

Previous studies have shown that anemones can be important predators in marine communities (Kruger & Griffiths 1997; Posey & Hines 1991) and as such the impact of S. ornata as a predator on the sandy-shore communities within Langebaan Lagoon is of interest. The coelenteron content analysis revealed that S. ornata ate a very limited diet, sand-dwelling polychaetes and amphipods being the most important prey items. The diet of an anemone normally reflects the faunal communities supported in its habitat (Chintiroglou & Koukouras 1992; Kruger & Griffiths 1998) and as the sandy-shore communities provide habitat for sand-dwelling and burrowing species (Branch et al. 2010), it may explain their importance in the diet of S. ornata. Sea anemones are sessile passive feeders mostly relying on prey to come to them (Sebens 1981) and have various feeding mechanisms to meet their daily nutritional requirements. These include absorption of dissolved organic matter, forming of mutualisms with photosynthetic zooxanthellae (Tsuchida & Potts 1994), filter-feeding (Lampitt & Paterson 1987) and feeding macrophagously on either motile or wave-tumbled prey (Sebens 1981). Although the feeding mechanisms of S. ornata have not been investigated, it occurs in a sheltered habitat with little wave action, a habitat type where anemones rely to a great extent on mobile prey (Kruger & Griffiths 1998). It is therefore likely that S. ornata employs a prey catching strategy, as is common to anemones in temperate regions (Kruger & Griffiths 1996). The feeding mechanism of an anemone is an important determinant of its impact on the surrounding communities. Anemones relying on wave action to supply them with prey would not have direct impacts on communities, as prey items would already have been removed from source populations (Kruger & Griffiths 1998). In contrast, anemones such as S. ornata that rely on mobile prey are more likely to affect their communities as prey are actively removed from communities.

Diet analyses can be affected by a variety of factors, such as poor preservation of specimens, egestion (Lampitt & Paterson 1987), digestion rate and turnover times (Kruger & Griffiths 1997), as they influence the retention time and quantities of prey items available for analysis (Kruger & Griffiths 1998). In addition, diet can vary on a seasonal and temporal basis (Kruger & Griffiths 1996). Thus, the present study cannot be considered an exhaustive analysis of S. ornata's diet. However, it does provide an indication of the taxa consumed and insight into the predatory role of S. ornata.

Although the ecological role of S. ornata has not been documented in its native range, it was anticipated that this anemone would negatively affect native communities as this is a well-documented impact associated with many sandy-shore invaders (Grosholz et al. 2000; Robinson et al. 2007; Ross et al. 2004). As predicted this pattern was observed for S. ornata, with invaded areas supporting a higher invertebrate abundance, biomass and diversity, as well as altered communities. As these changes were not driven by the loss of the species that were found to dominate S. ornata's diet, it is unlikely that this anemone is exerting its impact on native communities through its role as a predator. Rather, these changes may be the result of impacts on the invaded habitat by S. ornata and indirect effects on native biota. This anemone appears to alter the surrounding habitat, consolidating sand and trapping coarse sediment (Robinson pers. obs.). This has been observed elsewhere where alien species have changed soft sediment systems by altering abiotic factors such as water movement and sediment characteristics (Berkman et al. 2000; McKinnon et al. 2009; Ruiz et al. 1997). In turn, sediment type and structure influence the distribution, abundance and biomass of associated species (Nel & Branch 2013; Thrush et al. 2003). This may explain the higher invertebrate abundance, biomass and elevated diversity observed in the sites where S. ornata has invaded. Notably, the high abundance of tanaids recorded in the presence of S. ornata is likely a reflection of domination by coarser sediments, as these crustaceans are known to have reduced rates to tube building in the presence of sediments below 63 µm (Krasnow & Taghon 1997). In addition, tanaids themselves are known to change community structure by altering the habitat and preying on small and juvenile soft-bodied invertebrates (Oliver & Slattery 1985). It is thus suggested that a synergy between the primary impacts of S. ornata on habitat structure and the secondary impacts of these small crustaceans ultimately result in the observed differences between invaded and uninvaded areas.

This study aimed to provide an updated baseline on the status and distribution of S. ornata, to assess the diet of this alien anemone and to investigate its impact on the structure of the sandy-shore communities of the WCNP. A survey of the lagoon found that S. ornata remained restricted to the WCNP, although its distribution had changed. Notably this anemone was recorded in N. capensis seagrass beds for the first time. Nonetheless it is unlikely that S. ornata will spread extensively within this marine protected area because of the restricted nature of the habitats in which it occurs. Diet analysis revealed the dominant prey items consumed by S. ornata were indigenous sand-dwelling polychaetes and amphipods. Multivariate analysis showed that in invaded areas this alien altered sandy-shore community structure, increased diversity, density and biomass of native invertebrates. Despite these impacts, the geographically restricted nature of the habitat types currently invaded by S. ornata limits the potential of this alien to negatively affect indigenous biota within the WCNP. However, routine monitoring of surrounding areas should be undertaken as the occurrence of S. ornata on rocky shores and in kelp beds in its native range suggests that this anemone could spread to outside the park.