Translocation of marine species is one of the main processes altering the oceans globally (Ruiz et al. 1997, 1999, Geburzi and McCarthy 2018). Certain human activities favour the artificial dispersal of marine species. In particular, maritime traffic is the main dispersal vector of species outside their native ranges, involving ca. 93161 vessels yearly (Molnar et al. 2008, UNCTAD 2017). In a single day, ca. 10000 marine species are transported in the ballast waters, or as a part of the “biofouling” attached to hulls, across the world’s oceans (Carlton 1999, de Castro et al. 2017). The translocation of species underpins the concept of introduced species or non-indigenous species (NIS), i.e. species transported by humans, accidentally or deliberately, outside their natural distribution ranges, even across biogeographical barriers (Thomsen et al. 2015).

Most introduced species fail in their initial attempt to settle in new habitats; some NIS can adapt to new conditions, but fail to proliferate and alter recipient native communities (Geburzi and McCarthy 2018). A small group, however, can severely impact native biodiversity and communities, and have flow-on social and economic effects on fisheries, aquaculture and tourism (European Commission 2014, Thomsen et al. 2015); these are considered invasive species. Typically, invasive species have large reproduction, growth, dispersal and colonization rates. Some of them can be even considered “ecosystem engineers” because of the large physical and biological implications of their presence (Cuddington and Hastings 2004). In general, invasive species are a major threat to local biodiversity and a serious environmental concern for the conservation of the oceans, altering the functions and services delivered by local ecosystems (Bax et al. 2003, Thomsen et al. 2010, 2015).

In the scientific literature, there is growing evidence of the number of invasive species, covering a broad range of taxonomic lineages and affecting almost all eco-regions of the world (Thomsen et al. 2015). Within the phylum Cnidaria, only a small number of potential invasive hydroids (class Hydrozoa) have been described (González-Duarte et al. 2016a). This is the case of Pennaria disticha Goldfuss, 1820 in Hawaii (Miglietta et al. 2015), Turritopsis dohrnii (Weismann, 1883), considered as a global invader (Miglietta and Lessios 2009), Maeotias marginata (Modeer, 1791) in the Baltic Sea (Väinölä and Oulasvirta 2001), Moerisia lyonsi (Boulenger, 1908) in North America (Ma and Purcell 2005), and Blackfordia virginica Mayer, 1910. The last species is found in estuarine zones of all oceans (Mills and Sommer 1995, Haydar 2012), exerting a high predation pressure on planktonic crustaceans and fish eggs, with effects on commercial fish populations (González-Duarte et al. 2016a). Similarly, Clytia hummelincki (Leloup, 1935) is invasive in the Mediterranean, preying on eggs and fish larvae (González-Duarte et al. 2016b). Importantly, the arrival and establishment of many NIS species into new locations has been typically described from just one nearshore habitat, e.g. of 327 non-native marine and estuarine species reported in North America, 46% were reported only on hard substrata, and 22% occurred on both hard substrata and soft sediments (Ruiz et al. 2009).

Macrorhynchia philippina Kirchenpauer, 1872 (Hydrozoa, Alglaopheniidae) has a circumglobal distribution in tropical, subtropical and temperate waters of the planet (Ansín Agís et al. 2001, Watson 2002). In the last few years, the species has reached new areas, particularly at temperate latitudes, facilitated by its high dispersal capacity (Çinar et al. 2006, Morri et al. 2009) and by global warming (Gravili et al. 2013). In some places, e.g. the eastern Mediterranean, the species is considered invasive (Çinar et al. 2006, Zenetos et al. 2010). In the eastern Atlantic, it was initially recorded at Cape Verde Islands (Ritchie 1908) and on the coasts of Guinea Bissau (Billard 1931). It was later found in the archipelagos of Madeira and Selvagens (Bianchi et al. 1998, Ansín Agís et al. 2001, Wirtz 2007). Molecular analyses further suggested that the species was introduced through shipping, either in the ballast waters or attached to ship hulls (Moura et al. 2012). In the Canary Islands, the species was first recorded by Riera et al. (2016).

M. philippina is a large-sized (20-30 cm in height) colonial hydrozoan, which can become locally abundant under favourable environmental conditions (Moura et al. 2012). Colonies have a bushy appearance, resulting from the irregular ramification of the main branches arising from a central, dark-brown, axis (Fig. 1). From the main branches, fine ramifications support rows of white polyps. The colonies are very stinging and could be harmful to humans due to the presence of large nematocysts (100 μm in length, Marques et al. 2002); indeed, this species has been given the common name “stinging hydroid”. The increase in abundance of this species could reduce tourism activities and have a negative impact on local economies (Çinar et al. 2006, González-Duarte et al. 2016a).

Most works on NIS merely report the presence of these species outside their native ranges of occurrence (Ruiz et al. 1999). For many marine NIS, the mode of introduction, establishment success across habitats and impacts on marine ecosystems remain largely unknown (Streftaris et al. 2005, Ojaveer et al. 2015). In this context, to obtain a deeper understating of potential invasions, it is necessary to describe the ecological pattern, e.g. abundance and size structure patterns across scales of temporal and spatial variation (Byers et al. 2002, Streftaris et al. 2005). In this study, to the best of our knowledge we describe, for the first time, the process of colonization of an oceanic island by M. philippina. We describe the abundance and size structure patterns of the species on three nearshore habitats (rocky reefs, seagrass meadows and rhodolith seabeds) before, during and after its demographic explosion.

On both rocky reefs and rhodolith seabeds, abundances increased dramatically in 2017 in comparison with 2014, but the increase was lower in seagrass meadows (‘Habitat × Year’, p=0.0018, pairwise tests, Table 2, Fig. 3). Overall, abundances were greater (from 0 to a maximum of 138 colonies/100 m²) in 2017 than in 2016 and 2014 (‘Year’, p=0.000001, Table 2). Within each habitat, we detected large inter-site variation in the abundance of colonies (‘Site(Habitat)’, p=0.000001, Table 2, Fig. 3), i.e. abundances were greater at certain sites. The size structure of colonies varied between habitats (χ²=158.17, df=18, p<2.2e-16). On seagrass meadows, the colonies were smaller than on reefs and rhodolith seabeds (Fig. 4).

This work has demonstrated that Macrorhynchia philippina underwent a widespread demographic explosion in Gran Canaria Island after the first sighting in 2015 (Riera et al. 2016). The colonies were identified near (<10 nautical miles) two industrial ports on Gran Canaria Island (industrial ports of Las Palmas and Arinaga). Most likely, the species arrived as biofouling attached to ship hulls (Riera et al. 2016). A similar pattern has also been observed at Tenerife Island, where colonies were observed near two industrial ports (Santa Cruz de Tenerife and Granadilla) (A. Brito and L. Moro, pers. com.). Fouling species can be easily dispersed by maritime traffic (Ferreira et al. 2006), and oil rigs in particular can transport a wide range of biological assemblages (Wanless et al. 2010). The arrival of NIS at the Canary Islands has increased dramatically since 2011, when the traffic of oil rigs and digging vessels rose, particularly at Gran Canaria and Tenerife Islands. Certain NIS that are considered invasive have been recorded since then, including the corals Oculina patagonica and Tubastraea coccinea (Brito et al. 2017) and the portunid crab Cronius ruber (González et al. 2017). Similarly, the observation of dozens of tropical fish species in the vicinity of industrial ports in Gran Canaria, particularly around the industrial port of Las Palmas, has been linked to the maritime routes of oil rigs (Brito et al. 2011, Triay-Portella et al. 2015, Pajuelo et al. 2016). The progressive warming of waters around Gran Canaria Island in the last few decades may have also facilitated the successful colonization of this hydroid, which has an affinity for warmth (Gravili et al. 2013).

In just two years, the hydroid underwent a rapid expansion, colonizing large areas of the east coast of Gran Canaria Island. Importantly, the species has been identified in a range of habitats (rocky reefs, rhodolith seabeds and seagrass meadows; this study), as well as artificial structures (shipwrecks and fishing traps; Riera et al. 2016). On some reefs, there has been a massive increase, i.e. from 0 to 138 colonies/100 m². This pattern is common during the initial phases of invasion by invasive NIS (Thomsen et al. 2015), i.e. introduction-lag time-expansion (Geburzi and McCarthy 2018). Such invasion success suggests that M. philippina has a large colonization potential and ecological plasticity, i.e. the species was found in varying habitat types. In addition to a large dispersal capacity, this species displays a high investment in growth of somatic tissues and a high competition capacity for space (González-Duarte et al. 2016a). Considered a ‘broadcast spawner’, this hydroid has a complex life cycle. Colonies are hermaphroditic, and male and female medusoids are released by gonozoids inhabiting the same colony (Bourmaud and Gravier-Bonnet 2004). Overall, after fertilization, the hydrozoan planula larvae settle in less than 24 h, suggesting low dispersal capacities over distances of no more than a few kilometres (Calder 1993, Postaire et al. 2017). However, this seems to be sufficient to colonize wide areas at small oceanic islands such as Gran Canaria Island.

In this study, we detected between-habitat differences in the degree of establishment of colonies. Larger abundances were observed in rocky reefs and rhodolith seabeds than in seagrass meadows. The main ecological drivers affecting the biology of hydroids are the substrate type, light, hydrodynamics, salinity, sedimentation, temperature, food availability and pollution (Boero 1984, Gili and Hughes 1995). Benthic colonies are attached to either natural or artificial hard bottoms; the stability of the substrate is key for the attachment, growth and survival of colonies. The greater stability provided by rocky bottoms explains the greater abundance of colonies on reefs. Rhodoliths, however, can be sporadically overturned during strong currents or stormy episodes (Marrack 1999, Basso et al. 2009, Vale et al. 2018); this may help to explain the comparatively lower abundance of colonies there. Finally, seagrass meadows are very unstable, because there is a lack of hard structures, so colonies were only found attached to shells, isolated pebbles or unburied seagrass rhizomes (pers. obs.). In addition, the continuous movement of seagrass canopies and sand may hinder the settlement of larvae. In turn, only a few species of hydrozoans inhabit soft bottoms (Gili and Hughes 1995).

Water movement is a key element affecting the size of the colonies of hydroids. As they are passive filter-feeders, there is a trade-off between sufficient surge for the delivery of food and strong flows detaching the colonies from the substrate (Gili and Hughes 1995). According to these authors, the size of colonies tends to decrease with increasing surge intensity. In our study, smaller colonies were observed on rhodolith seabeds than on rocky reefs. The former habitat, in the study area, is under the influence of strong tidal currents (Otero-Ferrer et al. 2019), which fits the observation of Gili and Hughes (1995). On the other hand, in the study area, seagrass meadows are always located in semi-enclosed areas protected from strong currents and swells (Fabbri et al. 2015). A comparatively lower intensity of water flows most likely supplies a lower amount of food resources for the colonies of the hydroid, which could explain the smaller size of colonies in this habitat than on the reefs and rhodolith seabeds.

Invasive NIS tend to affect recipient assemblages (Carlton 2009). The effects of the invader can be classified as positive (e.g. habitat creation, increased species richness, provision of food) or negative (e.g. competition, habitat destruction, decreased species richness) (Thomsen et al. 2015). This study, however, has not analysed potential effects on native marine flora and fauna; at present, we cannot speculate on potential ecological implications.

In summary, this study has described the colonization process of a colonial hydroid at an oceanic island. The successful establishment of M. philippina demonstrates its great ecological plasticity. Continuous monitoring of coastal habitats is necessary to address the implications of the arrival of invasive species such as this hydroid in the study region. Urgent environmental policies to prevent, track and mitigate the arrival of NIS is required in this regard, particularly in the context of intense traffic of oil rigs and drill ships in the study region.