Introduction

Several years after its accidental introduction into the Black Sea in the early 1980s, the ctenophore, Mnemiopsis leidyi A. Agassiz, formed massive blooms that coincided with abrupt decreases in zooplankton, ichthyoplankton, and fisheries (Vinogradov et al., 1989; Vinogradov & Shushkina, 1992; GESAMP, 1997). Since the 1980s, the introduced distribution of M. leidyi spread to the Sea of Azov (1988), the Sea of Marmara (1989–1990), and the northern Aegean Sea (1990) (Shiganova et al., 2001). Additional introductions have transported the ctenophore to the Caspian Sea (1999) (Ivanov et al., 2000; Shiganova et al., 2001) and the North and Baltic seas (first observations 2006) (Faasse & Bayha, 2006; Javidpour et al., 2006). Within the Mediterranean Sea, M. leidyi rapidly spread from the Aegean Sea to adjacent waters of Turkey and Syria (Kideys & Niermann, 1994). More recently, M. leidyi was reported from the Northern Adriatic Sea and France (Shiganova & Malej, 2009), Israel (Galil et al., 2009), and Italy (Boero et al., 2009).

Mnemiopsis leidyi is a very versatile species native to estuaries and coastal regions along the eastern coasts of North and South America (GESAMP, 1997). Within its native range it occurs in coastal waters with temperatures ranging between 2 and 32°C and salinities of <2–38 (Purcell et al., 2001). It has flourished in non-native waters with winter temperatures above 4°C, estuarine salinity, and high productivity (e.g., Shiganova et al., 2001). The extraordinary success of M. leidyi in the Black, Azov, and Caspian seas has been attributed to the lack of predators in combination with the deteriorated conditions for fish populations in those waters due to eutrophication, pollution, and over fishing (e.g., Purcell et al., 2001; Oguz, 2005). The northern European coastal waters, where M. leidyi is considered to be established (Javidpour et al., 2008), share the above characteristics.

The first report of M. leidyi in the Mediterranean Sea was in the Aegean Sea, where gelatinous predators (e.g., Beroe spp. ctenophores) occur, and temperatures and salinities are much higher than in other non-native locations; M. leidyi appeared to have limited success there, and the establishment of persistent populations was uncertain (Kideys & Niermann, 1993; Shiganova & Malej, 2009). Short-term occurrences in Mediterranean ports were attributed to ballast-water transfers from the Black Sea to those ports (summarized in Galil et al., 2009).

Siapatis et al. (2008) developed a predictive model to identify the potential habitat of M. leidyi in the Mediterranean basin. Their model shows potential habitat around the entire Mediterranean coast. In each of 3 years modeled from environmental conditions (2004–2006), the Catalan and Levantine coast of Spain and the Levant (eastern Mediterranean) areas including Israel showed high probabilities of M. leidyi occurrence. In 2009 shortly after publication of that prophetic paper, M. leidyi was reported along Israeli shores (Galil et al., 2009), in the Adriatic Sea (Faris, 2009), and from the Catalan and Levantine coast in Spain (Fuentes et al., 2009).

Here, we report high abundances of M. leidyi in coastal waters of Israel and Spain in winter–summer 2009, in conjunction with hydrographic conditions and the abundances of zooplankton and scyphomedusan species. We compare these factors in 2009 with those in 2008 when M. leidyi was not widely reported. We also present the first molecular confirmation of the identity of M. leidyi in the Mediterranean Sea, along with possible source region(s).

Materials and methods

Field sampling

Spain

Beginning in 2000, semi-quantitative data were compiled on the occurrences of large gelatinous plankton, predominantly the scyphomedusae, Pelagia noctiluca (Forskal), Rhizostoma pulmo (Macri), and Cotylorhiza tuberculata (Macri), along the Catalan coast. The abundances of jellyfish on the beaches (without distinguishing among species) were classified in three categories: 1 (few, <10 animals per beach), 2 (many, <1 medusa m−2), and 3 (a lot, >1 medusa m−2) (Gili et al., 2007, 2008). This classification system was retained for observations that began in 2007 when the “Medusa Project” began monitoring the presence of individual jellyfish species along the Catalan coast.

The “Medusa Project” monitoring comprised a network of organizations covering the entire Catalan coastline. The Medusa Project is financed by the Catalan Water Agency (ACA) with the main goals of studying and monitoring the jellyfish proliferations along the Catalan coast of the NW Mediterranean Sea. The ACA records the relative abundance (few, many, a lot) of jellyfish daily at more than 300 beaches covering the 69 municipalities of Catalonia. Data are recorded by beach inspectors who observe what is on the sand and in the near-shore water at fixed locations and by inspectors in boats 200-m offshore beyond the swimming area. The project also involves participation of the Emergency Services from 26 of the 69 municipalities that each completes a form daily to report the relative abundance of jellyfish at the beaches (ACA, 2009). The Fisherman Associations of Catalonia also report to the Medusa Project. All the collected information, including daily observations of jellyfish from ACA boats, beach inspectors, and emergency services, are summarized on the ACA web page and are available to the public (ACA, 2009). As part of the observation and monitoring protocol, data on water temperature, transparency, salinity, chlorophyll a, and nutrients have been recorded since 2000 along the coast.

As part of a cooperative project, information about jellyfish presence in other parts of Spain is also available. In Denia (Levantine coast of Spain, south of Catalonia), daily visual inspection of the beaches has been conducted by the local Administration (Environmental Department) and by its summer beach emergency services (Red Cross) since summer 2001. The presence of jellyfish is recorded daily as part of this monitoring procedure. Water samples have been collected monthly since 2001 at four locations, separated by 3–5 km, along the coast at 4–5 m depth and 300–700 m from shore. Data on water temperature, O2, pH, and nutrients were recorded in those samples.

Israel

Jellyfish occurrences in coastal or offshore waters of Israel have been recorded on an ad hoc basis; many of the observations were made by fishermen. Data on the occurrences and abundances of medusae and ctenophores in 2009 were compiled during trawling (fishing) cruises and from interviews with trawl, seine, and inshore fishermen at the ports of Jaffa and the Kishon River (Haifa). Trawlers used 40–48-mm stretched diamond mesh nets with a vertical opening of 2–3 m, trawling at depths of 15–100 m. Inshore and purse-seine fishermen trawled at 0–50 m depth with 22–120-mm (mostly 50–60 mm) mesh gill- and trammel-nets with a 12-m vertical opening in shallow waters and up to 24 m in deep waters. Purse-seiners used 18–24-mm mesh nets.

Presence/absence and the relative abundances of gelatinous animals were recorded. The abundances of gelatinous zooplankton reported were “none,” “occasional” (sporadic sightings in surface waters or 1–5 specimens in each trawl), or “abundant” (either swarms in surface waters or >5 specimens in each trawl). Zooplankton abundances in coastal waters were determined between March and May 2009 by means of horizontal net tows at 5-m depth for 3 min at 1 m s−1 and vertical tows from the seafloor to the surface at 1 m s−1 with 64- and 303-μm plankton nets. Surface seawater temperatures were recorded at Ashdod Port during the past 12 years (CAMERI, 2008) and both surface seawater temperatures and salinities were recorded near Ashkelon (IEC, 2009) during 2009.

Molecular analysis

Ctenophores collected in the field were provisionally identified morphologically as M. leidyi based on a typical morphotype, namely that the lobes extended to the region of the statocyst (see Faasse & Bayha, 2006). In order to confirm field identification, a few specimens of this consistent morphotype were collected, preserved, and identified molecularly. Molecular species identification was performed making use of the DNA sequence from the nuclear internal transcribed spacer (ITS) region (ITS-1, 5.8S, ITS-2), a region that has been useful for ctenophore identification in the past (Podar et al., 2001; Bayha et al., 2004; Faasse & Bayha, 2006). A total of three ctenophores were collected and preserved in 75–100% ethanol, including one from the vicinity of Haifa, Israel (HAF-1) and two from the vicinity of Salou, Spain (SAL-1 and SAL-2). DNA extraction was performed using a standard CTAB protocol (Dawson et al., 1998), and polymerase chain reaction (PCR) amplifications were performed using an Applied Biosystems 2720 Thermal Cycler. The entire ITS region was PCR amplified using primers KN-8 (ATTACGTCCCTGCCCTTTGTA) and KN-9 (GCAATCCCAAACAGTCCGACTCTTC) in conditions consisting of 95°C for 3 min, followed by 38 cycles of 95°C for 45 s, 52°C for 1 min, 72°C for 1.5 h, and a single cycle of 72°C for 10 min (then 4°C). Successful PCR products were treated with Exonuclease I and Shrimp Alkaline Phosphatase (USB Corp.) to prevent interference by unused primers and dNTPs in sequencing reactions. PCR products were then directly cycle sequenced by Northwoods DNA, Inc. (Solway, MN) using primers KN-8 and KN-11 (ATTTGAGCTGTTCCCTGTTCGT) (Bayha, 2005). All sequences were deposited into GenBank. All DNA sequences were assembled using Seqman II (DNAStar, Inc.), and electropherograms were amended by eye for bad sequence calls. Loci that showed clear peaks of two different bases were coded as degenerate (i.e., Y = C and T). Sequence identity was evaluated by performing BLASTN searches against GenBank (Altschul et al., 1997).

Results

Field sampling

Spain

During the summer of 2009, we received many reports from the entire length of the Catalan coast describing the presence of an unusual jellyfish species, which we identified as M. leidyi (Fig. 1). Only the observations for which we visually confirmed that the reported jellyfish species was M. leidyi are given in Table 1. Many other similar reports were received from several additional locations along the Catalan coast, but we were not always available to make identifications; therefore, M. leidyi may have been even more commonly distributed than reported here. The confirmed reports show that large swarms of M. leidyi were observed from the beginning of July to the end of September along the entire coast from Cap de Creus in the northern Catalan coast to Alcanar in the south. The presence of M. leidyi also was confirmed in the south of Catalonia in Denia (Valencia, Levantine coast of Spain) and Cabrera (Balearic Islands) (Table 1; Fig. 1). Most of the M. leidyi observations made by ACA during summer 2009 were in the highest abundance category 3.

Fig. 1
figure 1

Reports of blooms of Mnemiopsis leidyi in the Mediterranean Sea in 2009 in Spain, Israel, and Italy

Full size image
Table 1 Records of Mnemiopsis leidyi along the Spanish NW Mediterranean coast during summer 2009
Full size table

The surface coastal waters of the Catalan Sea (NW Mediterranean) during the summer months of 2009, when the presence of M. leidyi was confirmed (from July to September), were characterized by salinities between 30.5 in the south, where the delta of the Ebro River is located, and 38.2, and temperatures between 21.5 and 25.8°C. Temperature and salinity (S) of the waters where M. leidyi was present ranged from 22.7 to 25.3°C and 37–37.9 S (Table 1).

Along the coast of Denia, a high-density area of M. leidyi was detected from 16 to 18 July 2009, with densities ranging from 5 to 100 ind. m−2. The ctenophores were seen only on those days in an area of about 2,500 m2 at depths of 0–4 m (Table 1). The high-density patch dispersed after 18 July; however, low densities of M. leidyi were observed by SCUBA divers between June and August in that area. Temperatures of coastal waters in Denia (2 m depth) ranged from 12.5°C in March to 27.5°C in July. In this area, the effects of river runoff are very important; for example, nitrate values are normally very high, reaching a maximum of 361 μg l−1 in November 2006.

The observations of large jellyfish (scyphomedusae and hydromedusae) were atypical along the Catalan coast during the spring and summer months of 2009. The typical pattern in 2000–2008 showed that the greatest jellyfish abundances occurred during July and August (Gili et al., 2007, 2008); however, in 2009, most jellyfish occurred during May and June (Fig. 2). In July, most of the jellyfish species usually seen then (P. noctiluca, Aequorea forskalea Péron & Lesueur, and R. pulmo) nearly disappeared, which coincided with the period when M. leidyi was most abundant and frequently observed (Fig. 2).

Fig. 2
figure 2

The numbers of beaches with scyphozoan and hydrozoan jellyfish during the beach season 18 May to 30 September 2008 (circles) and 2009 (squares). The arrows mark the period of time when high numbers of Mnemiopsis leidyi ctenophores were observed in 2009 in the Catalan Sea, Spain

Full size image

Israel

Observations, including relative abundances, of gelatinous zooplankton in Israeli coastal waters were recorded from January to July 2009 with an emphasis on medusae (mainly Rhopilema nomadica Galil, Spanier & Ferguson and R. pulmo) and ctenophores (Table 2). This record-keeping was stimulated by the appearance in January of a massive bloom of M. leidyi, which had not been observed prior to this year in such abundances in these coastal waters. The earliest reports of the ctenophore swarms were from purse seine and trawl fishermen, describing “marmalade” or “medusa jelly” that made it difficult to lift the fishing gear. Continuous onboard observations of the ctenophores led to subsequent visual confirmation during a SCUBA diving expedition to 32°51′02″N–034°56′33″E. All of the ctenophore swarms were observed within 5 nautical miles of shore. Ctenophores were reported to occur either sporadically or abundantly from January to mid-June (Table 2). Most of the medusae observed in 2009 were R. pulmo, which was often abundant in January through March, and several patches of R. nomadica that occurred in June (Table 2). Routine coastal monitoring did not reveal unusual patterns in seawater temperatures, salinities (39.1–39.8), or chemistry during spring and early summer 2009 as compared with other years (Herut et al., 2005, 2006, 2008); however, winter 2008/2009 temperatures were higher than average winter temperatures (Fig. 3).

Table 2 Observations of ctenophores in coastal waters of Israel during 2009
Full size table
Fig. 3
figure 3

Sea water temperature (°C) of last 12 years from Palmachim, Israel. Solid triangles maxima, open triangles minima, crosses means. Solid circles mark the 2008–2009 winter–spring water temperatures

Full size image

Molecular species identification

The sequences for the nuclear ITS region (ITS-1, 5.8S, and ITS-2) from all three ctenophores sequenced were identical in length (638 base pairs) and nearly identical in sequence to published M. leidyi sequences, indicating positive identification as M. leidyi. Sequences for the three ctenophores analyzed differed by at most a single degenerate base (T in SAL-1 and HAF-1 and Y [C and T] in SAL-2). All sequences were extremely similar to published M. leidyi sequence for ITS (NCBI Accession #AF293700), all being the same length and differing by 3 (SAL-1 and HAF-1) or 4 (SAL-2) bases out of 638 (all degenerate changes, i.e., T vs. Y, etc.), indicative of within-species differentiation. In comparison, the three ctenophores differed markedly from the next closest match (Bolinopsis sp.—Accession #BSU65480) in being 2 bp shorter and diverging by 19–20 bp.

Discussion

Arrival and transport of Mnemiopsis leidyi in the Mediterranean Sea

Although May 2009 was the first confirmed report of M. leidyi on the Spanish coast, this species was present before but remained unrecorded until its populations increased. We located unpublished information about the presence of M. leidyi in Cabrera (Balearic Islands) (OCEANA, 2008); we confirmed its presence there in 2009 (Fig. 1). Earlier occurrences of M. leidyi in other Mediterranean areas may be established by examination of formalin-preserved plankton samples, as in Purcell (1988). High numbers of M. leidyi along the Mediterranean coastlines of Israel (Galil et al., 2009), Italy (Boero et al., 2009), and Spain during summer 2009, plus the fact that the species was also present in Spanish waters in 2008, strongly suggest that its population is established in the Mediterranean.

Molecular species identification confirmed representative ctenophores as M. leidyi through the use of DNA sequence from the nuclear ITS region. Some previous articles on invasive ctenophores in the Baltic Sea apparently have misidentified the ctenophores to the genus Mnemiopsis (Kube et al., 2007; Lehtiniemi et al., 2007); the errors only were recognized upon the use of molecular identification tools (Gorokhova et al., 2009).

Although the molecular dataset is small (n = 3), it may also give some preliminary indication of the original source region(s) of some of the ctenophores. Two of the sequenced ctenophores (SAL-1 and HAF-1) contained an ITS composite genotype that was previously found in invasive M. leidyi from the Black Sea (southwestern Black Sea and Gelendzhik Bay, Russia) and the Sea of Azov (various locations), as well as in native ctenophores from the United States (only from the Gulf of Mexico at Mobile, Alabama and Atlantic Ocean at Miami, Florida) (Bayha, 2005), possibly indicating common recent ancestry for these animals. The third ctenophore (SAL-2) contained an ITS composite genotype not previously found in any geographic region (Bayha, 2005). These results must be interpreted cautiously, however. First, the dataset is extremely small, so the findings are specific only to these individuals and do not reflect anything regarding the greater Mediterranean population (i.e., others may have originated elsewhere). Description of the entire population would require the examination of significantly more Mediterranean individuals. Secondly, while the ITS region historically has been useful at the species level, the ITS genetic region occurs in multiple copies (Arnheim et al., 1980) and its use in population-level studies can be problematic (Vollmer & Palumbi, 2004). Therefore, ITS data ideally should be corroborated with data from additional single-copy genetic marker(s). Nevertheless, our preliminary data are consistent with two of our specimens having a common source with Black and Azov sea invaders, as well as ctenophores from the Gulf of Mexico (Mobile, Alabama) and/or Miami (Florida) in the native range.

The introduction of M. leidyi and its distribution throughout the Mediterranean Sea probably resulted from transport by both ballast waters from the Black Sea region and currents. M. leidyi was first reported in the Mediterranean Sea in 1990, transported into the Aegean Sea by water flow through the Dardanelles from the Sea of Marmara or by ballast water from ships. The ctenophore was reported in Turkey and Syria soon thereafter (1992 and 1993; Table 3). No other reports were made in the Mediterranean until 2005, when M. leidyi appeared in France and 2006 in the Gulf of Trieste (Table 3). The simultaneous blooms of M. leidyi all along the Mediterranean coasts of Israel, Spain, and Italy in 2009 (Fig. 1; Boero et al., 2009; Fuentes et al., 2009; Galil et al., 2009) suggest that such wide-spread, abundant ctenophores could not have been distributed by either currents or shipping in 2009, but that the ctenophores already were present and were stimulated to bloom in favorable conditions in 2009.

Table 3 Hydrological characteristics and mesozooplankton biomass (μg C) or abundance (number m−3) of the seas of the Mediterranean basin, and the year when Mnemiopsis leidyi was first reported and the seasons when it is known to be active in each location
Full size table

Transport of M. leidyi in ballast water to Spain could have been to Barcelona, which was listed as the Mediterranean port with greatest ship traffic in 2006 (REMPEC, 2008). Of the Spanish locations with confirmed M. leidyi, only Denia represents an important commercial and passenger port; however, because all ships use the container system to transfer cargo, they do not release ballast water; therefore, it is unlikely that M. leidyi arrived in Denia by ship transport. The ctenophores found in the southern Catalan Sea as well as in Denia may be explained by circulation patterns in those areas.

The general circulation pattern in the Catalan Sea is characterized mainly by a current that follows the continental slope, in geostrophic equilibrium with a shelf/slope hydrographic front (Font et al., 1988). This current, the Liguro-Provençal-Current (LPC), is directed from the northeast towards the southwest partially governed by the effect of a permanent cyclonic gyre north of the Balearic Islands at the western side of Gulf of Lyon. This current transports the major outside Mediterranean surface waters to the Catalan Sea from the Gulf of Lyon along with considerable runoff from the Rhone River. This mesoscale circulation pattern persists throughout the year but is most intense in winter and spring (Font et al., 1988). In the southern Catalan Sea, the LPC follows the shelf break, entering the Gulf of Valencia and continuing to the Eivissa Channel (Salat, 1995). Water balances in the region show that the volume of water flowing southward through the Eivissa Channel is always smaller than the transport along the shelf break by the LPC and Atlantic waters that penetrate into the eastern Catalan Sea by the Eivissa Channel (García et al., 1994). The southern penetration of Atlantic waters mainly occurs through the channel between Eivissa and Mallorca all year, but this water mass remains in the eastern sector of the region (Monserrat et al., 2008). In years with mild winters, the flow of Atlantic waters through the Eivissa Channel is more intense especially in late spring and summer. These circulation patterns make intrusion of Atlantic waters into the Catalan Sea from the northeast with surface waters more probable than flow through the eastern Balearic Islands and the Gulf of Lyon. However, after mild winters, the intrusions from the south through the Eivissa Channel could had been of great importance incorporating Atlantic waters flowing around the Iberian Peninsula from the Strait of Gibraltar, being this another possibility for the introduction of M. leidyi into the southern Catalan Sea waters.

Hamad et al. (2005) analyzed satellite thermal images of Mediterranean surface waters and proposed a revision to the accepted scheme of large-scale flow in the Mediterranean Sea. Hamad et al.’s analysis indicates that there is a direct link between water flowing from the Aegean Sea, where M. leidyi has occurred since 1990, to the Levant. Alternatively, because the swarms of M. leidyi were observed mainly in the southern half of Israel, between Ashkelon and Netanya and considering that the predominant current in this region is along shore from south to north (Rosentraub & Brenner, 2007; Zviely et al., 2007), it is possible that M. leidyi was released in ballast water from ships traveling toward the southern Israeli port of Ashdod or to Egypt.

Factors affecting the success of Mnemiopsis leidyi in the Mediterranean Sea

Environmental conditions

Conditions for M. leidyi in the Mediterranean are quite different from the conditions in most native and introduced habitats. M. leidyi typically inhabits shallow estuaries and coastal waters in its native American waters (reviewed in Kremer, 1994; Purcell et al., 2001). In its northern temperate habitats (Rhode Island and Maryland), temperatures range from 1°C in the winter to 30°C in the summer and salinities range from ≤2 to 32. The annual peak in biomass of northern M. leidyi populations is in mid-summer to fall, which have occurred earlier with rising temperatures since 1950 (Sullivan et al., 2001; Costello et al., 2006a). Ctenophore biomass declines in the autumn when temperatures decrease and the ctenophores overwinter in nearshore bottom-water refugia (Costello et al., 2006b). Ctenophores in more southerly native locations (Florida and Texas) experience higher temperatures (winter minima of 7–18°C and summer maxima 31–32°C) and higher salinities (14–45) than in the north. The populations peak in winter when temperatures are lower (~20°C) than the rest of the year, and the population can display several small peaks (20–30 ml m−3 live volume) throughout the year.

These same patterns are seen among the locations where M. leidyi was introduced earlier. The Black, Caspian, North, and Baltic seas in the northern range are often estuarine, with a summer bloom of M. leidyi that subsides in cool temperatures (reviewed in Purcell et al., 2001; Shiganova et al., 2001). The southern introduced habitats in the Sea of Marmara and Mediterranean Sea are similar to the southern U.S. native habitats, with high salinities and temperatures (Shiganova et al., 2001; Shiganova & Malej, 2009). The only study of osmoregulation in ctenophores showed M. leidyi to conform to salinity within the range tested (8–23; Foshtomi et al., 2007). Clearly, the habitats of M. leidyi seem unrestricted by most physical conditions, with the exception of temperatures <4°C in high latitudes.

Water conditions along the Catalan coast are suitable for M. leidyi throughout the year (Fig. 4), because they would not be limited by cold winter temperatures as in northern latitudes. M. leidyi is present all year in the Aegean Sea (Shiganova et al., 2004), but its seasonal distributions in other Mediterranean locations are unknown.

Fig. 4
figure 4

Monthly measurements of sea temperature (°C) and salinity in the surface waters of the Catalan Sea from March 2008 to September 2009. Solid triangles salinity, crosses temperature

Full size image

Food resources

Zooplankton abundance directly affects ctenophore abundance. M. leidyi was found in waters with high zooplankton stocks and did not occur where zooplankton biomass was <3 mg C m−3 (reviewed by Kremer, 1994; Purcell et al., 2001). In native locations, peak zooplankton biomass (~100 mg C m−3) was higher in northern than in southern latitudes (~25 to 50 mg C m−3).

In contrast to rich environments where M. leidyi is found, the north western Mediterranean is considered to be an oligotrophic sea, with low nutrient concentrations but relatively moderate levels of pelagic primary production (100–150 g C m−2 y−1; reviewed by Estrada, 1996). In particular, the Catalan Sea is a highly dynamic oligotrophic environment, subjected to important forcing from the alternation of stratification and mixing periods and from the strength of mesoscale singularities (Saiz et al., 2007). Two features of special importance for productivity in the Mediterranean are the deep chlorophyll maximum (DCM) at the base of the pycnocline during most of the stratification period, and the occurrence of two phytoplankton blooms (Estrada, 1996), which are followed by two peaks of zooplankton in late winter–spring and in autumn. During stratification, the DCM is an oasis for zooplankton feeding and accentuates the difference between the poor surface layer (0.2–0.3 μg chl l−1) and richer waters at depth (1–2 μg chl l−1) (Estrada, 1996). During the stratified period, the trophic web based on the microbial loop could assume great importance; microzooplankton is responsible for most of the secondary production in warm waters (Fernández de Puelles et al., 2007; Saiz et al., 2007). In the Mediterranean Sea, there appears to be strong coupling between the microbial food webs and the upper trophic levels (Cushing, 1989). During the warm months, the diverse mesozooplankton community is predominated by filter-feeding organisms that efficiently consume the small fractions of the food web (Calbet et al., 2001).

Mnemiopsis leidyi is very versatile in the types of prey it consumes, which range from microplankton to mesozooplankton and ichthyoplankton; M. leidyi larvae ≤5 mm diameter contained diatoms, dinoflagellates, and ciliates, and larger larvae also contained copepod nauplii and copepodites (reviewed in Purcell et al., 2001; see also Sullivan & Gifford, 2004; Rapoza et al., 2005). Several studies report inverse correlations of M. leidyi and zooplankton abundances (reviewed in Purcell, 1988); when M. leidyi were abundant, clearance rates of the ctenophores were sufficient to reduce zooplankton populations (reviewed in Purcell et al., 2001; Shiganova et al., 2001; see also Purcell & Decker, 2005; Riisgård et al., 2007). In addition, M. leidyi had higher clearance rates on copepods than did scyphomedusae of equal biomass (Purcell & Decker, 2005). The wide variety of prey types and high feeding rates could have allowed this species to establish in the oligotrophic conditions of the NW Mediterranean.

Other important food resources for M. leidyi are fish eggs and larvae. In the NW Mediterranean, ichthyoplankton has a pronounced seasonal variability because most fishes (neritic, small pelagics, and migratory species) spawn during spring and summer. The most abundant pelagic species are European sardine and anchovy, which have non-overlapping spawning periods, autumn–winter and spring–summer, respectively (Sabatés et al., 2007). Consumption of ichthyoplankton by M. leidyi can be very important in native and non-native environments (reviewed in Purcell et al., 2001; Shiganova et al., 2001); however, the importance of M. leidyi predation on zooplankton and ichthyoplankton in the recently invaded Mediterranean regions are not yet known.

In addition to the wide spectrum of prey that this ctenophore can exploit, the success of introduced M. leidyi also could be attributed to reduced competition from over-fished populations of zooplanktivorous fish (Oguz et al., 2008; Siapatis et al., 2008). In the NW Mediterranean Sea, over-fishing is believed to be one reason that gelatinous zooplankton seem to be increasing in abundance in recent decades (Gili & Pagés, 2005). Two important features of the fishing activity in the Mediterranean Sea are the multi-specificity of catches and the absence of large single stocks, especially in the demersal regime, as compared with those that inhabit other seas (Coll et al., 2006). The development of fishing technologies and over-capitalization, with an increasing demand for marine resources, is placing intensive pressure on marine resources in the western basin, and the general assessment suggests that most demersal stocks are fully exploited, while some pelagic stocks are even overexploited (Bas et al., 2003; Leonard & Maynou, 2003; Coll et al., 2006).

It is not clear what lead to the bloom of this ctenophore in the Levant but it may be related to the dietary similarities of larval fish, zooplanktivorous fish, and M. leidyi. The fishing pressure along Israeli coasts continues to be very high (Pisanty & Grofit, 1991) and may have caused an untimely imbalance in favor of the invasive ctenophore, enabling it to flourish, despite the relatively low abundances of micro- and mesozooplankton detected in the coastal waters of Israel during the period that M. leidyi was present. Zooplankton abundances in the eastern Mediterranean Sea off the coast of Israel generally are very low, and median values ranged from 256 to 1,261 m−3 for microzooplankton and 13–196 m−3 for mesozooplankton, as recorded from March to May 2009.

Predators and competitors

Interannual variation in M. leidyi abundance is strongly related to predator abundance in their native waters (reviewed in Purcell et al., 2001; Purcell & Decker, 2005). A variety of predators occur, including the ctenophore, Beroe ovata Brugière, the scyphomedusae, Chrysaora quinquecirrha Desor and Cyanea capillata (Linnaeus), harvestfish, Peprilus alepidotus (Linnaeus), and butterfish P. triacanthus (Peck), but their effects on M. leidyi populations have rarely been studied. Low and high salinity waters can serve as a refuge for M. leidyi from predators, as in Chesapeake Bay. Purcell et al. (2001) concluded that the key difference between native U.S. habitats of M. leidyi and the Black, Azov, and Caspian seas was the lack of predators there, perhaps enabling the devastating blooms of the ctenophore in those non-native habitats. Subsequently, B. ovata arrived in the Black and Azov seas and has limited M. leidyi populations since 1999.

The NW Mediterranean Sea has a diversity of potential predators of M. leidyi. The ctenophores Beroe cucumis Fabricius and Beroe forskalii Milne Edwards are common (Shiganova & Malej, 2009); however, Beroe spp. normally are abundant only in spring and early summer along the Catalan coast (F. Pagès, pers. com.), suggesting that they were not abundant there when M. leidyi occurred in high numbers (July and August 2009). Nevertheless, at the beginning of July 2009, both M. leidyi and P. noctiluca were present in the coastal waters and P. noctiluca ate M. leidyi in the laboratory (V. Fuentes, pers. obs.). In late July 2009, P. noctiluca almost disappeared from the Catalan coast; their absence coincided with high abundances of M. leidyi (Fig. 2). Another abundant potential predator is the hydromedusa, A. forskalea; its congeners are voracious predators of gelatinous species (Purcell, 1991).

In addition to the potential predators, many potential competitors for zooplankton foods occur in the NW Mediterranean, including the large jellyfish, P. noctiluca, C. tuberculata, R. pulmo, and A. forskalea, and many species of small hydromedusae and siphonophores (Sabatés et al., this volume), as well as several species of zooplanktivorous fishes (Leonard & Maynou, 2003; Sabatés et al., 2007).

Jellyfish observations in Israel indicate that in addition to the persistent swarms of R. nomadica and R. pulmo that have been documented along the Mediterranean shores since the 1980s (Galil, 2007), there are occasional blooms of novel exotic gelatinous species that had not been previously recorded. A comprehensive list of scyphomedusae of the Mediterranean coast of Israel includes P. noctiluca, Aurelia aurita, C. tuberculata, Phyllorhiza punctata von Lendenfeld, Cassiopea andromeda (Forsskal), R. pulmo, and R. nomadica (in Galil et al., 1990). Despite annual outbreaks of stinging jellyfish, few attempts have been made to quantify these swarms (Spanier & Galil, 1991; Lotan et al., 1992) or to follow their spatio-temporal dynamics. As in the NW Mediterranean, all of these species are potential competitors with M. leidyi for zooplankton prey, and P. noctiluca is a likely predator of M. leidyi. The importance of these various species as predators and competitors of M. leidyi in Mediterranean waters remains to be investigated.

The lack of basic information on these exotic invasives throughout the Mediterranean led to the 2008 CIESM initiative, “JellyWatch” (http://www.ciesm.org/marine/programs/jellywatch.htm) and the “Medusa Project” (ACA, 2009) created to document the frequency and distribution of jellyfish blooms in the Mediterranean Sea. Those efforts need to be expanded to include other gelatinous groups in addition to large jellyfish. For example, the records for ctenophores in the Mediterranean Sea are scarce; M. leidyi may have been resident previously but was not noticed before the spectacular 2009 blooms.

Conclusions

Nearly simultaneous blooms of the invasive ctenophore, M. leidyi, occurred throughout the Mediterranean Sea in 2009. This voracious zooplanktivore has caused wide-spread ecosystem disruption and damage to fisheries following its accidental introduction to the Black Sea basin in the early 1980s. It subsequently invaded waters of the North and Baltic seas, where its spread and abundance may be restricted by cold temperatures and native gelatinous predators. Conditions in the Mediterranean Sea are very different (comparatively warm temperatures, high salinities, and low productivity) from those in the northern habitats invaded by M. leidyi; therefore, M. leidyi is unlikely to be constrained by physical conditions in the Mediterranean Sea. On the other hand, a variety of gelatinous species are potential predators and competitors of M. leidyi that may constrain its populations in the Mediterranean. Whether or not M. leidyi flourishes, the presence of this species will change the Mediterranean ecosystems and poses a threat to fisheries.