Fouling community dominated by Metridium senile (Cnidaria: Anthozoa: Actiniaria) in Bahía San Julián (southern Patagonia, Argentina)

1 Instituto de Ciencias del Ambiente, Sustentabilidad y Recursos Naturales (ICASUR), Unidad Académica San Julián, Universidad Nacional de la Patagonia Austral, Colón y Sargento Cabral, (9310) Puerto San Julián, Santa Cruz, Argentina. E-mail: martin_jpablo@yahoo.com.ar 2 Instituto de Investigaciones Marinas y Costeras (IIMyC), Facultad de Ciencias Exactas y Naturales – CONICET, Universidad Nacional de Mar del Plata, Funes 3250, (7600) Mar del Plata, Argentina.


INTRODUCTION
Studies of biofouling communities in Argentinian harbours began in the 1960s and have provided a fairly complete picture of the fouling communities from Buenos Aires Province and North Patagonia (Bastida 1971, 1973, Bastida et al. 1980, Martin et al. 2000. These studies have mainly dealt with faunal inventories, settlement cycles of species, trophic relationships, epibiosis, resistance of organisms to control biofouling substances and other general ecological topics (Bastida et al. 1976(Bastida et al. , 1980. They helped to clarify the mechanisms of colonization of these communities, as well as the processes of ecological succession on artificial substrates in marine and estuarine environments. More recently, studies were carried out in central Patagonia, specifically in the port of Comodoro Rivadavia (Rico and López Gappa 2006). However, studies of biofouling communities are virtually non-existent in southern Patagonia.
One of the objectives of studies on biofouling communities is to attempt to solve problems related to the bio-deterioration of man-made structures in the marine environment caused by organisms. These include docks, boats, facilities such as refrigeration plants and power stations that use seawater as a coolant, and underwater archaeological sites. Knowledge of the composition and succession cycle of fouling communities allows different strategies to be developed to avoid the settlement of marine organisms on artificial surfaces, or to control excessive development of communities (Bastida et al. 2004, Dürr andThomason 2010). On the other hand, studies of fouling communities in harbours that are situated within, or adjacent to, reserve areas often have other objectives, usually related to the conservation of biodiversity. Knowledge of fouling communities settled in relatively pristine areas can establish a baseline for monitoring the impact on marine biodiversity caused by urban, port and industrial activities and detecting entry and establishment of invasive alien species related to shipping traffic (Lewis and Coutts 2010). This is the case of Bahía San Julián, which is part of a protected area within which intertidal communities have recently been studied, whereas the subtidal community is practically unknown. Knowledge of the community is particularly important in this case, as in 2013 local port activity began to recover slightly and fishing boats from harbours located northwards of the locality began to operate.
The sea anemone Metridium senile is a cosmopolitan species, predominantly found subtidally and frequently reported in biofouling communities of cold-temperate areas in the Northern Hemisphere. It often acts as a structuring species of the community by affecting the early stages of ecological succession (Nelson and Craig 2011). Riemann-Zürneck (1975) and Zamponi and Acuña (1991) reported this species in different parts of the Argentinean Continental Shelf at depths of 95-120 m. Subsequently, the species was also reported from the port of Cape Town in South Africa (Griffiths et al. 1996, Acuña andGriffiths 2004), where it was possibly introduced by shipping traffic from European ports. It is important to note that currently there is an ongoing study of different populations of M. senile worldwide to find out whether they are different species or subspecies (Häussermann and Försterra 2009).
In Argentina, the importance of this species in the benthic community of coastal southern Patagonia remains unknown but few studies have been done on biofouling in the region. The main goals of this study were to provide information on a fouling community dominated by M. senile in the harbour of Bahía San Julián, and to analyse processes of epibiosis that this anemone exerts as a strategy of colonization leading to its dominance in the community. This study also aims to provide a baseline regarding the characterization of the fouling community of Bahía San Julián, depending on the possible revival of the local fishing port and the potential risk of introduction of exotic marine species.

Study area
Bahía San Julián is located on the central coast of Santa Cruz Province (southern Patagonia, Argentina) between 68°50'01" and 67°35'50"W and 49°00'12" and 49°47'02"S ( Fig. 1). It is a deep inlet of the sea about 20 km long and 8.8 km in maximum width, with a relatively narrow mouth (approximately 700 m wide). The Bay has a maximum depth of 35 m and includes a wide shallow terminal area (sac). It receives cold waters of sub-Antarctic origin and others with relatively low salinity from the Patagonian Coastal Current, a branch of the Malvinas Current. The salinity of water inside the Bay is between 33 and 34 and temperatures vary between 5°C (winter) and 14°C (summer). The tidal regime is of a semidiurnal macrotidal type, with maximum amplitude of 8.93 m and an average of 6.15 m (Falabella et al. 2009, SHN 2009.
The harbour of the city of Puerto San Julián, with a population of about 12000 people, is located on the northwest coast of the Bay. This has a dock comprised of a concrete manoeuvring platform 62 m long supported by cylindrical steel tubes. This harbour experienced increased activity in the 1990s and early 2000s, involving mainly fishing vessels but also merchant vessels. However, the port activity there was never very great, being always lower than that of other harbours of the region. In 2000, for example, only four fishing vessels were registered as operating, according to the report of authorities of Puerto San Julián. Occasionally, during this period foreign flag vessels captured by authorities of the Argentina Naval Prefecture for illegal fishing operations also arrived. In 2007 repairs of the wharf commenced, with a view to the future revival of the harbour, but by 2009, when the samples of this study were taken, the port was still virtually inactive. Moreover, it is important to mention that Bahía San Julián integrates a protected area, which also includes the Cormorán and Justicia Islands that are found inside, under a protection regime of limited use. Additionally, the Peninsula of San Julián was declared as a provincial reserve (Chebez 2005). The coast of the Bay is dominated by plains formed mainly by mud (silt + clay) and fine sand, with large extensions of intertidal mudflats on the upper levels. The mudflats are inhabited by saltmarshes of Sarco cornia perennis, and infaunal communities dominated by polychaetes and small bivalves, such as Darina solenoides and Mysella patagona in the middle and lower intertidal. Some sectors with a higher proportion of gravel enable the development of intertidal mussel beds of Perumytilus purpuratus, Mytilus edulis platensis and Aulacomya atra atra (Sar 2010, Ituarte et al. 2012. Hard substrates are scarce inside the Bay and are formed by sedimentary rocks. These areas are inhabited by intertidal communities comprised of P. purpuratus, M. edulis platensis and A. atra atra. These substrates are more abundant and achieve greater development in the open sea coast adjacent to the Bay, sustaining a major intertidal benthic community characterized by the development of mussel banks.

Sampling and statistical analysis of the community
In December 2009 several steel piles supporting the wharf of Puerto San Julián were extracted for repair and refurbishment. These operations allowed us to carry out the first study characterizing the fouling macrobenthos community of the lower intertidal and subtidal inside the Bay. Sampling was conducted using a framework of 20×20 cm and metal spatula, obtaining four randomly distributed replicas from each of three different levels: Level 1, corresponding to lower intertidal; Level 2, corresponding to the subtidal between 3 and 4 m depth; and Level 3, corresponding to the subtidal between 6 and 7 m depth (both from mean low spring tide). The samples were fixed in 5% formalinseawater solution and transported to the laboratory for analysis. Organisms were identified to the lowest taxonomic level possible and counted under a stereoscopic microscope, then preserved in 70% alcohol. Abundance data were analysed by multivariate analysis using clustering (hierarchical agglomerative clustering) and ordination techniques (non-metric multidimensional scaling, MDS), using the Plymouth Routines in Multivariate Ecological Research (PRIMER) statistical package version 6.1 (Clarke and Warwick 2001). The data were previously transformed using square root to down-weight the effect of dominant species, and then the Bray-Curtis similarity index was applied. The statistical significance of the groups of samples was tested by analysis of similarities (ANOSIM permutation test), in order to evaluate the hypothesis of differences in community composition between depth levels. The similarity percentages (SIMPER) routine was then used to identify the main species that characterize the community at each depth level (Clarke and Warwick 2001). For each sample we calculated the number of species S; the diversity index of Shannon-Wiener H'= -Σp i log(p i ), where p i is the proportion of the total number of individuals that belong to species i; and Pielou evenness J=H'/log S. The abundance data of M. senile and diversity of the samples were compared between sampling levels by ANOVA and post-hoc LSD test, after verifying homogeneity of variance using Levene's test (Zar 1996). The significance level used in all tests was p<0.05.

Analysis of Metridium senile epibiosis
Settlement on other organisms and sedentary polychaete tubes, when they were inhabited, was considered epibiosis. For these studies we identified and counted all organisms that served as substrate for Metridium senile, and also the number of individuals of the sea anemone epibionts on each substrate-organisms. Epibiosis frequency (%F) for each substrate-species in the sample was calculated as %F i =N i /N t ×100, where N i is the number of individuals of the i species upon which M. senile epibiosis was found, and N t is the total number of individuals of all species in the sample with epibiosis of M. senile. Furthermore, the intensity of M. se nile epibiosis on each substrate-species was calculated as the total number of individuals of M. senile fixed on all individuals of the species in the sample. Spearman rank correlation analysis between the abundance of each substrate-species in the sample and the number of individuals of the substrate-species with epibiosis of M. senile was conducted to determine whether there was a relationship between these variables. Subsequently, we determined whether there were significant differences between the relative abundance of species-substrate in the sample and epibiosis frequency (%F) using the test Chi 2 for contingency tables 2×2, in order to determine if M. senile had higher affinity for any particular substrate-species. The average number of anemones epibionts per individual of the different substrate-species was calculated and compared by Kruskal-Wallis and Mann-Whitney tests (Zar 1996). The significance level used in all tests was p<0.05.

Community composition and abundance of Metridium senile
The fouling community was dominated by Met ridium senile in all the three studied levels. This species was found together with different groups of macrozoobenthos organisms, varying in importance in the community according to sampling depth (Figs 2 and 3). Table 1 lists all taxa identified in the community, with their respective observed average abundances per level.
At Level 1 (lower intertidal), M. senile had an average relative abundance of 43%, and was accompanied by the mussels Perumytilus purpuratus (17.5%) and Aulacomya atra atra (12.9%) among the most abundant species. Density of M. senile at this level reached 12900 ind. m -2 with an average of 10406 ind. m -2 .  senile settled on the mussels Mytilus edulis platensis and Aulacomya atra atra at Level 1. C, M. senile settled on mussels and the sea squirt Paramolgula gregaria at Level 2.
The diversity of the community peaked at Level 1, significant differences were observed with respect to Level 2 (ANOVA F 2,9 =5.06, p=0.03, LSD Test 1>2, p=0.0124), and the lowest value was recorded at Level 2, coinciding with the lowest number of species in the community and the greater abundance of M. senile and M. insidiosum, which together accounted for over 83% of total faunal abundance (Fig. 5).
The two-dimensional MDS plot showed a clear separation of samples by depth level (Fig. 6). The Table 1. -List of taxa found in the fouling community of Bahía San Julián and their average abundances for depth level. In cases of Porifera and bryozoans their abundance are indicated with "+" signs, each "+" indicating its presence in one replica (from one to four).

Metridium senile epibiosis
Metridium senile settled directly to the surface of the piles and upon different sessile or low-mobility organisms present in the community, as well as on structures of biological origin, such as sedentary poly-  chaete tubes. In Level 1, M. senile was epibiont on five different species. The substrates with the highest frequency of epibiosis (%F) at this level were the mussels Perumytilus purpuratus and Aulacomya atra atra, which accounted for 98% of the biological substrates utilized by the sea anemone (Figs 3 and 7). In subtidal levels, the richness of biological substrates increased significantly. At Level 2, the number of biological substrates used was 12, the highest frequency of epibiosis being upon A. atra atra (52%) and Paramolgula gregaria (13%) (Figs 3 and 7).
At Level 3, epibiosis on seven organisms was observed, with the greatest frequency on Crepipa tella dilatata (37%), Sabellidae polychaete tubes (38% overall) and P. gregaria (13%), see Figure 7.  The abundance of different substrate-species in the samples showed a significant correlation with the number of individuals of each species with epibiosis by M. senile (r s =0.51, t (87) =5.51, p<0.0001) (Fig. 8). Moreover, no significant differences between the relative abundance of substrate-species in the samples and epibiosis frequency (%F) indicated that, in general, M. se nile showed no preference for any particular substrate. Significant differences were found only for Austro megabalanus psittacus at Level 1 (χ 2 =7.39, p=0.006). In this case, the barnacle comprised an average relative abundance of 14% among the substrate-species of M. senile, but had an epibiosis frequency lower than 1%, indicating that there was a negative selection of M. se nile for this substrate at this depth level.
Epibiosis intensity of M. senile (Fig. 9) at Level 1 was higher on the mussels P. purpuratus and A. atra atra. In the latter species up to 138 sea anemones were recorded on the total specimens in a sample. At Level 2, the highest intensity of epibiosis was observed on A. atra atra, P. gregaria and M. edulis platensis. At this level the intensity of epibiosis on A. atra atra reached 570 individuals of M. senile. At Level 3, epibiosis intensity was higher on P. gregaria, with a maximum of 198 individuals of the sea anemone recorded.
The average number of anemones epibionts per individual of substrate-species (Fig. 10) at Level 1 was highest on the mussels A. atra atra (0.81 epibionts per individual) and M. edulis platensis (0.5 epibionts per individual). At Level 2, the highest number of epibionts per individual was observed on P. gregaria, A. atra atra and M. edulis platensis, with averages of 25.5, 15.5 and 14.5 respectively. At Level 3, the number of epibionts per individual was higher on P. gregaria, with an average of 18 anemones recorded per host.

DISCUSSION
The fouling community from Bahía San Julián showed similar traits in its composition to those found in natural rocky substrates in the region, with a high abundance of mussels in lower intertidal levels, from where they were progressively replaced by other species in the subtidal. In the Ría Deseado (200 km N of Bahía San Julián), the association of Perumytilus purpuratus, Mytilus edulis platensis and Aulacomya atra atra dominates and characterizes the community in the lower intertidal level, mainly in the vertical rocky substrates inside the ría. In the shallow subtidal, the community of Ría Deseado is characterized by a decrease in the abundances of P. purpuratus and M. edulis platensis and a predominance of A. atra atra and Polyzoa opuntia, accompanied by Paramolgula gregaria, Crepidula dilatata dilatata (=Crepipatella dilatata), Photinula caerulescens, Balanus psittacus (=Austromegabalanus psittacus), Aerothyris venosa (=Magellania venosa) among the most conspicuous species (Ringuelet et al. 1962, Otaegui and Zaixso 1974, Zaixso and Pastor 1977. Despite the similarities found in the community of vertical hard substrates between the two sites, the absence of the sea anemone M. senile in works of the 1960s, 1970s and early 1980s in the Ría Deseado is striking. Indeed the very low abundance of Actiniaria as a group in the subtidal of Ría Deseado (Callebaut Cardu and Borzone 1979) is very  different from the situation found in the present study, in which Actiniaria, represented only by M. senile, was a dominant group. Moreover, López Gappa et al. (1982) recorded the following actiniarians among fauna associated with Macrocystis pyrifera holdfast in Ría Deseado: Isoedwardsia nidarosiensis, Aiptasiamorpha minima, Acontiophorum mortenseni and Isanthus cap ensis. All of these showed very low abundance in the community.
The only exotic species of the fouling community of Bahía San Julián identified was the gammarid amphipod Monocorophium insidiosum, which is characterized as invasive on the coast of Argentina. This species, usually seen in high densities in biofouling of harbours, is widely distributed in the world, being transported mainly by human action. In Argentina, it was detected in the harbour-fouling community of Mar del Plata in the 1970s, when it was identified as Coro phium comparable with Corophium insidiosum. Subsequently, it was reported for other harbours of Buenos Aires Province and the Patagonian coast (Bastida et al. 1980, Alonso de Pina 1997, Martin et al. 2000, Orensanz et al. 2002. The virtual absence of exotic species in fouling and natural hard substrate communities of Bahía San Julián is in marked contrast with fouling from other harbours of the region, such as Puerto Madryn, Puerto Deseado and Comodoro Rivadavia (Fig. 1), which have high shipping activity. In these harbours several invasive alien species have been reported in recent years, such as Undaria pinnatifida, Carcinus maenas, Ascidiella aspersa, Balanus glandula and Cryptosula pallasiana (Piriz and Casas 1994, Orensanz et al. 2002, Hidalgo et al. 2005, Rico and López Gappa 2006, Schwindt 2007, Martin and Bastida 2008. However, none has been found so far in Bahía San Julián. The low shipping activity in the local harbour, which has been practically inactive for several years, could be responsible for the absence of exotic species common in other harbours of the region. The absence of these alien species is consistent with the assumption that the dispersion along the Patagonian coast is fundamentally linked to commercial shipping traffic (commercial fishing and transport) and the passive spread by currents has limited scattering power due to the large extension of the coast. Thus, this study provides interesting information on this problem and provided a baseline for Bahía San Julián, especially taking into account the possible reactivation of port activities in the locality.
A particular feature of the fouling community of Bahía San Julián is the dominance of the sea anemone Metridium senile at all levels sampled. This funding has not been reported so far in the intertidal or subtidal natural hard substrates from the region. However, M. senile is currently a common species in hard substrate communities from the open sea around Bahía San Julián, along with other species of sea anemones, such as Antholoba achates and Bunodactis octoradiata (Garese et al. 2014). This sea anemone forms aggregates in free lower intertidal substrates, usually in tidal pools and on horizontal and vertical walls, and to a lesser extent as an epibiont (Acuña et al. 2011). This suggests that protected conditions in the inner of Bahía San Julián and the existence of a limited, but stable artificial substrate generate a suitable environment for successfully settlement of a sea anemone population, reaching high abundances and dominating the community. A similar phenomenon was observed in harbour biofouling of cold-temperate areas in the Northern Hemisphere (Nelson and Craig 2011).
It is important to note that M. senile is also currently an abundant species within the Ría Deseado, where it mainly inhabits the lower intertidal and shallow subtidal rocky substrate (Acuña et al. 2011). In that location, it can also be observed in intertidal shores composed of fragments of rock and mud inside the port and surrounding areas, and as an epibiont of the seaweeds Un daria pinnatifida, Macrocystis pyrifera and Lessonia sp. (Martin unpublished data). However, as mentioned above, the species was not cited in previous works in that locality in the 1960s 1970s and early 1980s (Ringuelet et al. 1962, Zaixso 1975, Zaixso and Pastor 1977, Callebaut Cardu and Borzone 1979, López Gappa et al. 1982, which suggests that its entry to the Ría Deseado could have occurred from the mid-1980s or 1990s, when commercial shipping traffic and fishing activity started and intensified in that locality. It is also important to note that the existence of this species could not be confirmed along the Chilean coast, despite extensive samplings conducted in the last decade (Häussermann andFörsterra 2005, Häussermann 2006).
At San Julián harbour, shipping activity reached its peak during the 1990s, when fishing vessels operated from Argentinian harbours located northwards (e.g. P. Madryn and C. Rivadavia), and from some from foreign harbours. During the 1990s, for example, vessels from Spain and Cape Town (South Africa), where M. senile has been recorded as an invasive alien species since 1995 (Griffiths et al. 1996), were recorded as entering Bahía San Julián. Given the cosmopolitan nature of M. senile, its high abundance at present, and the fact that it has not been cited in previous works on the region, we recommend granting this anemone the status of an alien invasive species in Argentina, or at least the rank of a cryptogenic species: one for which it is not possible to demonstrate whether it is native or introduced (Carlton 1996).
There are several factors that make M. senile the dominant species of the biofouling community of San Julián harbour, one of which is its great ability to bind to different substrates, including other sessile organisms or structures derived from them, as in the case of polychaete tubes. The analysis of epibiosis processes shows that the biological substrates commonly used by M. senile are molluscs, in particular the mussel A. atra atra, the gastropod Crepipatella dilatata, and the tunicate Paramolgula gregaria. However, this would not be a result of active selection of these biological substrates. The significant correlation observed between the abundance of substrate-species and the number of individuals of such species with epibiosis indicates that the frequency with which a given organism is used as a substrate for M. senile is determined largely by its abundance. Furthermore, statistical analysis makes it clear that there is no difference between the relative abundance of substrate-species and their frequency of use (%F), indicating that there is no selection of them by the anemone. Neither do the substrate-species exhibit effective mechanisms to prevent or inhibit the binding of the anemone, although, for example, cleaning behaviour to avoid epibiosis was described for Mytilus edulis by Theisen (1972) and Bruzzone (1982). Moreover, analysis of the attachment intensity and average number of epibionts per individual suggests that the number of anemones settled on a substrate species is also linked to the size of the substrate and the surface available for epibiosis. Thus, although the sea squirt P. gregaria is not an abundant species in the community, it is one of the most intensively used substrates in the subtidal levels, due to the large size of its specimens and hence the large surface available for settlement of the sea anemones. A. psittacus, was found to have a negative selection as a substrate at Level 1 because, despite its relatively high abundance in the community, the epibiosis frequency was very low on this organism, possibly because they were of very small size at this level (mean = 13.64 mm, sd = 5.76 mm basal diameter), providing a very small attachment surface. This was not the case at Level 2 where, though the abundance of barnacle is much lower than that at Level 1, the specimens were significantly larger (mean = 43.48 mm, sd = 3.68 mm basal diameter), offering a wider attachment surface for M. senile. In the latter case, no differences between the relative abundance of the barnacle in the community and the epibiosis frequency on the barnacle were observed.
We conclude that for M. senile the frequency and intensity of epibiosis is primarily determined by abundance and size of each substrate-species available. This underlines the high plasticity of the anemone to attach on several substrates, increasing their characteristic capacity of colonization. Thus, epibiosis emerges as a strategy of colonization when hard substrate availability is scarce, as is common for many benthic organisms (Wahl 1997), and it is particularly important for M. senile inside Bahía San Julián.
The high ability for epibiosis reinforces the potential of colonization of M. senile and, with other biological characteristics, allows us to understand its great adaptation to harbour environments and its cosmopolitan distribution. These features include the clonal mode of reproduction, sustained mainly by pedal laceration. This mode of asexual reproduction is typical of subtidal anemones and, according to the hypothesis of Sibly and Calow (1982), would be selected to promote growth in favourable environments such as the subtidal, while longitudinal fission would be the dominant mode in less favourable environments for growth, such as the intertidal zone. In the Northern Hemisphere (Gulf of Maine, USA), similar results for the same sea anemone were observed, where it was dominant in a community of subtidal rocky substrate and prevailed upon ascidians, bryozoans, sponges, coralline algae and other anthozoans (Sebens 1986). The aforementioned author highlights the rapid growth of the anemone and its dense aggregation, which excludes even colonial ascidians, as mechanisms that determine the competitive success of this species. It should also be noted that this anemone can develop tentacles with aggressive feature (catch tentacles), that may prevent or limit the settlement of other organisms, especially genetically different clones of the same species. In dense populations of M. senile, where essentially all space is occupied, the catch tentacles are lost because of "habituation" or "induced tolerance" to neighbours (Shick 1991).
In some subtidal communities in the Northern Hemisphere, mussels provide an important secondary substrate for populations of M. senile, showing coverage of up to 100% (i.e. Kaplan 1984). However, the mobility of mussel patch aggregates can frequently result in the death of their epibionts, which can be buried or located in unfavourable positions incompatible with life. Thus, the relative instability of the biological substrate, such as mussels, stimulates locomotion of the anemone causing the laceration of the pedal disc and the clonal reproduction of individuals, as observed by Kenneth et al. (1995). Moreover, experiments carried out by Nelson and Craig (2011) on artificial substrates show that the movements of M. senile kill new recruits from other fouling organisms by asphyxiation (smothering), possibly due to stress conditions generated by the low concentration of dissolved oxygen below the pedal disc. This process would allow M. senile to eliminate potential space competitors of surrounding substrate and continue dominating the space inside the fouling community. The dominance of M. senile observed in the fouling community of Bahía San Julián may then be explained by the effect on new recruits demonstrated by Nelson and Craig (2011) and also by its ability to settle on a wide range of biological substrates.