Marine Biology (2001) 138: 1195±1203 DOI 10.1007/s002270100539
M. Fine á H. Zibrowius á Y. Loya
Oculina patagonica : a non-lessepsian scleractinian coral invading the Mediterranean Sea
Received: 26 September 2000 / Accepted: 8 January 2001 / Published online: 15 March 2001 Ó Springer-Verlag 2001
Abstract The scleractinian coral Oculina patagonica De Angelis is a new immigrant from the Southwest Atlantic to the Mediterranean Sea, having established itself only recently along the Israeli coast. This species is the only scleractinian coral reported to have invaded a new region. In order to understand the swift establishment of this species along the Israeli coast, from 1994 to 1999 we studied its distribution, abundance, reproduction, recruitment, survival, and the eect of bleaching events on its population abundance. In addition, population studies of O. patagonica were performed at several localities along the eastern and western Mediterranean coasts. Highest abundance was recorded along the Spanish coast, reaching 307 colonies per 10 m line transect in shallow water. Second in abundance was the Israeli coast, with a maximum of 102 colonies per transect. O. patagonica was rare along the coast of Italy, and absent along the Mediterranean coast of France. During the study, both mortality and recruitment along the Israeli coast were very low. In contrast, recruitment along the Spanish coast was very high. Reproduction of the species was studied using gonadal histology. O. patagonica is gonochoric. Female gonads were ®rst observed in May and male gonads in July, both reaching maturity in late August and early September. Naturally occurring azooxanthellate colonies of O. patagonica inhabiting small dark caves developed gonads and spawned in parallel to zooxanthellate colonies exposed Communicated by O. Kinne, Oldendorf/Luhe M. Fine (&) á Y. Loya Department of Zoology, George S. Wise Faculty of Life Sciences and the Porter Super Center for Ecological and Environmental Studies, Tel-Aviv University, Ramat Aviv, 69978, Israel E-mail: m®
[email protected] H. Zibrowius Centre d'Oceanologie de Marseille (UMR CNRS 6540 DIMAR) Station Marine d'Endoume, rue Batterie-des-Lions, 13007 Marseille, France
to light. No gonads were found in zooxanthellate colonies that underwent bleaching during the reproduction season. The high incidence of bleaching events along the Israeli coast observed throughout the years of this study may explain the low recruitment of new colonies during the same period. In view of its current recruitment patterns, we expect further expansion of O. patagonica in range and abundance in the western Mediterranean, but very small expansion of the population in the eastern Mediterranean, due to repetitive annual bleaching events.
Introduction The zooxanthellate scleractinian coral Oculina patagonica (De Angelis D'Ossat 1908) was ®rst recorded in the Mediterranean Sea in 1966 in Savona harbor (Gulf of Genoa) (Zibrowius 1974). In 1973, it was ®rst noticed in southeastern Spain, in Alicante harbor (Zibrowius and Ramos 1983; Zibrowius 1992). At present it is known to inhabit many natural sites and harbors along hundreds of kilometers of coastline in southeastern Spain. Earliest sporadic records in the Levant area are from Egypt (el Agami) in 1981 and Lebanon (Khalde) in 1992 (Bitar and Zibrowius 1997). Along the Mediterranean coast of Israel, it was ®rst recorded in 1993 (Fine and Loya 1995). Major mechanisms of man-induced introduction of benthic marine species to a new area may include: (1) fouling on hulls of ships, (2) transport in ballast water, and (3) ``hitch-hiking'' through attachment to shell®sh and other commercially transported organisms (Carlton 1987; Carlton et. al. 1990; Zibrowius 1992). Transoceanic transport of O. patagonica as a fouling organism via the Straits of Gibraltar to the western Mediterranean (®rst Spain, then Italy) has been considered to be the most probable means of introduction (Zibrowius 1974, 1992; Zibrowius and Ramos 1983; Bitar and Zibrowius 1997). It is assumed that from the western basin, this successful invader subsequently spread to the Levant
1196
area via the intense intra-Mediterranean maritime traf®c. The distribution of a coral, however, can also be in¯uenced by its mode of reproduction. For example, hermaphroditism may in some cases prove to be more advantageous than gonochorism for successful colonization of a new area (Tomalison 1966); aggregated spatial distribution may be related to asexual reproduction (Carlon and Olson 1993). Most studies on stony coral reproduction have been performed on Indo-Paci®c and Caribbean reef corals (Harrison and Wallace 1990). Although coral reproduction was ®rst studied in the Mediterranean by Cavolini (1785, cited in De LacazeDuthiers 1873), surprisingly little research on both sexual and asexual reproduction of Mediterranean corals has subsequently been published. Asexual reproduction in scleractinians is diversi®ed and has a long record (Harrison and Wallace 1990). Kramarsky-Winter et al. (1997) described a new mode of asexual reproduction for O. patagonica: polyp expulsion, in which an individual polyp with its calice rises on an elongated calcareous stalk before detaching and settling elsewhere. When the new propagule has settled, its tissues spread down over the stalk and onto the substrate, and the polyp begins to form a new colony through budding. Polyp expulsion was observed in 19% of the investigated colonies in shallow water along the Israeli coast (Kramarsky-Winter et al. 1997). In reef corals, a variety of natural and man-induced factors are known to cause stress and to result in lowered fecundity (Harrison and Wallace 1990). One such factor is bleaching (see reviews by Glynn 1993; Brown 1997; Hoegh-Guldberg 1999), which in some cases terminates gametogenesis (Szmant and Gassman 1990). Bleaching of O. patagonica was ®rst recorded along the Israeli coast in the summer of 1993 (Fine and Loya 1995), and was explained by a bacterial infection to which high water temperature was a contributing factor (Kushmaro et al. 1996, 1997). In the present study we address the major intrinsic and extrinsic factors that may have contributed to the arrival and swift establishment of O. patagonica in the Mediterranean, including reproduction, size at reproductive maturity and growth rate. Concomitantly, we address the limiting factors, such as the bleaching events, that shaped the present population structure of O. patagonica along the Israeli coast.
Materials and methods Study sites In Israel the study was carried out from 1994 to 1999, at ®ve sites along 180 km of the Mediterranean coast of Israel (33°05¢N; 35°06¢E in the north to 31°40¢N; 34°32¢E in the south): RoshHanikra, Acre, Haifa, Sdot-Yam and Ashqelon (Fig. 1A). In southeastern Spain, three sites were surveyed from 1996 to 1998 (Fig. 1B): Calariona (1), Portman (2) and Alicante harbor (3). In Italy, three sites were inspected (Fig. 1B): Isola di Bergeggi (1),
Albisola marina (2) and Camogly (3). Observations were performed by skin diving and SCUBA. Distribution and abundance Average abundance and size of Oculina patagonica was calculated from 10-m line transects (Loya 1972). Colonies underlying the line were measured to the nearest milliimeter using a measuring tape. In addition, maximal length (L) and width (W) of each colony were measured, and the surface area (S, in mm2) was calculated using the formula S p
L W =42 . At each site in Israel, line transects (n=30) were placed at depths of 4, 10 and 30 m, and, at the Spanish sites, at depths of 1 and 6 m. Growth rate of O. patagonica was calculated by comparing monthly photographed color slides of 50 tagged colonies in Sdot-Yam (January 1994±October 1997). Since O. patagonica has a predominantly encrusting growth form (Fig. 2), the photographed colony area and annual growth rate were calculated to the nearest millimeter using a computerized image analyzer (Olympus CUE-3). Sample collection Samples for histology (fragments of 5±15 polyps per colony) were randomly collected at depths of 1±6 m (January 1994±October 1999) at all study sites. A total of 6±14 colonies of O. patagonica were sampled monthly to study the developmental stages of their gonads. From June to September, collecting was done every 2 weeks, and from May to September monthly samples were also collected from bleached colonies to compare their gonad development with that of healthy corals. In Marseille (France) monthly samples were collected from January 1995 to November 1999, from colonies experimentally transplanted (on various dates) from the Italian and Spanish sites (Zibrowius 1974, 1992). In addition, we collected samples from the Italian and Spanish sites for comparison with old samples (collected during the previous two decades by H.Z.), as well as recent samples from Lebanon and Egypt (Fig. 1B). All these samples (see Table 1) were used in the comparative histological study of the annual reproductive cycle. Environmental data In Israel, surface seawater temperatures (SSTs) were obtained using Onset Stow Away data-loggers. Salinity in tide pools was measured using an American Optical refractometer (model 10419). For the Marseille area, SSTs were provided by the ``Long Term Data Series'' measurements at the Station Marine d'Endoume in addition to in situ measurements taken during the sampling period. These Long Term Data Series include 30 min interval data, which illustrate the thermal eect of upwellings related to the Mistral wind. They enable study of the short-term water-cooling eect on gonad development. For Spain, monthly SSTs were obtained from the Comprehensive Ocean Atmosphere Data Sets (COADS) operated by the NOAA (National Oceanographic and Atmospheric Administration). Histology The samples were ®xed in 4% formaldehyde solution in seawater for 24 h, rinsed in fresh water and preserved in 70% ethanol. Decalci®cation was carried out using a solution of formic acid and sodium citrate for 15±24 h (see Rinkevich and Loya 1979). After decalci®cation, the tissue was rinsed in fresh water and transferred into 70% ethanol. Embedding of the tissue in paran was done using a Citadel embedding apparatus. Sections (6±10 lm thick) of polyps were stained with hematoxylin and eosin, and examined under a light microscope to determine the number and develop-
1197
Fig. 1A, B Oculina patagonica. Study sites along the eastern and western Mediterranean. A The Mediterranean coast of Israel. SdotYam was the main locality for studies on growth rate, mortality and recruitment. B General map of all study sites from east to west. Israel (see details in A), Lebanon, Egypt, Italy, France and Spain. The three sites along the Italian coast include: Isola di Bergeggi (1), Albisola marina (2) and Camogly (3). The three sites along the Spanish coast include: Calariona (1), Portman (2) and Alicante harbor (3) mental state of gonads. The diameters of all oocytes in a given polyp were measured, and their monthly mean growth was calculated. Spawning and larval development Every month between July and September colonies (n=20) were collected and transferred into aquaria at Tel-Aviv University. Each colony was kept separately in an aerated 5 l seawater aquarium, at a temperature of 26±28°C and under a 12 h light:12 h dark light regime. Water was replaced every week. When fully mature, the colonies were observed from sunset to 2:00 a.m. every night and
throughout the whole night during the full moon in September. Following spawning in aquaria, sperm and eggs were collected, mixed and allowed to develop in Petri dishes ®lled with ®ltered (0.2 lm) seawater. Development was monitored by pipetting out samples of ten embryos at intervals of 1 h for the ®rst 10 h following spawning, and examining them under a light microscope.
Results Distribution, colony size and growth We found Oculina patagonica to be abundant in wide areas along the Israeli coast at a depth range of 0.5±10 m. In addition to the natural habitats of sandstone reefs, this coral is abundant in disturbed sites such as areas exposed to domestic and industrial pollution, arti®cial boulders, submerged metal objects, jetties and marinas. Azooxanthellate colonies of O. patagonica were found in dark caves and crevices at a depth of 1±6 m. In tide
1198
Fig. 2 Oculina patagonica. An encrusting colony common on the reef ¯at and on vertical walls of the sandstone reefs o the coast of Sdot-Yam, Israel (50% of natural size) Table 1 Oculina patagonica. Sites and collection dates of samples for the comparative histological study of the gametogenic cycle Country
Site
Dates of collection
Israel
Ashqelon Sdot-Yam Haifa Acre Rosh-Hanikra Khalde Saadiyat Rmaili El Agami Alexandria Marseille Imperiaux de Terre Grand Conglu Albisola marina Isola de Bergeggi Camogli Portman Alicante harbor Cabo de Palos Cabo de Gata Punta el Santo Playa del Altet Isla del Descubridor
Jan 94±Dec 99 Jan 94±Dec 99 Jan 94±Dec 99 Jan 94±Dec 99 Jan 94±Dec 99 Jul 92 Sep 95, Oct 95, Nov 95 Jul 93 Nov 81 Nov 97 Jan 95±Dec 99 Jan 95±Dec 99 Jan 95±Dec 99 Jul 71, Oct 78, 95, Jul 96, Sep 96 Oct 78, Oct 95, Jul 96, Sep 96 Oct 97 Apr 81, Jul 82, Sep 96, 98 Apr 81, Jul 82, Sep 95, 96, 97, 98 Apr 81, Sep 96 Jul 82 Jul 82 Aug 82 Aug 82
Lebanon Egypt Francea Italy Spain
a Colonies experimentally transplanted to Marseille (on various dates) from the Italian and Spanish sites (Zibrowius 1974, 1992)
pools, colonies of this species are exposed to extreme temperature and salinity conditions. Temperature ranges from 10°C during winter to 40°C during the summer low tides, with daily amplitudes in spring and autumn reaching 9°C. Salinity in tide pools may range from 28& during heavy rain up to 50& at summer low tides. The mean abundance of O. patagonica colonies along the Israeli coast increased southward, from a minimum of 2.51 (n=10) colonies in Rosh-Hanikra (northern Israel) up to 102 (n=10) colonies in Ashqelon
Fig. 3 Oculina patagonica. Population abundance along the Israeli coast from north (Rosh-Hanikra) to south (Ashqelon)
(southern Israel) per 10 m line transect at a depth of 4 m (Fig. 3). At all sites, abundance decreased signi®cantly (P=0.001, paired t-test) below 4 m depth. At the Italian sites, O. patagonica populations were rare. Only one to seven colonies were recorded at the three study sites surveyed. In Portman, Calariona and Alicante harbor (Spain), abundance of O. patagonica per 10 m line transect was 205, 143 and 307 (n=10), respectively. Average colony size in Israel increased southward, from a minimum of 12262 mm2 (n=54) in RoshHanikra, at a depth of 4 m, up to 933737 mm2 (n=102) in Ashqelon, at a depth of 4 m (Fig. 4). In Portman, Calariona and Alicante harbor (Spain), average colony size was 648788 mm2 (n=200), 520480 mm2 (n=160) and 691314 mm2 (n=94), respectively (Fig. 4). Average linear growth rate recorded at Sdot-Yam (Israel) was 0.750.3 cm and 0.60.2 cm year±1 for colonies inhabiting vertical (n=138) and horizontal (n=55) substrates, respectively. Both mortality and recruitment were very low during the 3 years of observation. Out of 193 tagged colonies in the marked study site at Sdot-Yam (500 m2), 9 colonies died (predation and algal overgrowth), while 18 new colonies recruited to the area during 1995±1997. In Spain, however (e.g. Alicante harbor), many small colonies (less than ten polyps) were observed to inhabit the natural rocky substrate, shells of live gastropods of the immigrant species Crepidula calyptraeiformis, and trash (plastic bags and cans).
Reproduction O. patagonica was found to be a gonochoric coral. Out of >300 samples year±1 collected in the eastern and western Mediterranean during this study, the male/ female ratio was 1:1 at most of the sites, although in some cases, such as Savona (Italy), all colonies (n=7) were found to be females.
1199
Fig. 4A±H Oculina patagonica. Distribution of size groups along the Israeli (A±E) and Spanish (F±H) study sites
Colonies exceeding 2 cm in diameter (10±25 polyps) and estimated to be 1±2 years old, were already found to be fertile during the reproduction season. Female gonads ®rst appeared in May. Following a rise in seawater temperature, oocyte size increased. The primordial oocyte consisted of a slightly granular cytoplasm, with a large spherical nucleus. At maturity, ovaria ®lled the polyp cavity, especially its lower part. Maturation of gonads peaked in late August and early September, when oocyte mean size reached 100 lm (n=80). Each septum contained 10±25 oocytes (Fig. 5A). The mature oocytes were spherical, or sometimes irregularly shaped, containing an oval nucleus (20 lm in diameter) and a spherical nucleolus (5 lm in diameter) (Fig. 5B). The eggs did not contain zooxanthellae. Male gonads ®rst appeared in July, at the base of the polyp in the gastric cavity. The testes consisted of sacs of various sizes and shapes. In many cases various lobes of the same testis were found in dierent stages of development (Fig. 5C). The development of a male gonad started with a primary spermary, extending slightly from the mesentery into the cavity. Through successive divisions the primary spermary increased in size, gradually becoming ®lled with spermatogonia (Fig. 5D). Towards spawning, male gonads may reach a maximum of 300 lm (n=50) in diameter.
Spawning Simultaneous spawning of male and female colonies was observed in aquaria to last only two consecutive nights during the full moon of September (e.g. 20±21, 9±10 and 28±29 of September 1994, 1995 and 1996, respectively). Sperm was released in clouds (Fig. 5E) while eggs were released singly (Fig. 5F) and were negatively buoyant. Gamete release lasted for about 2 h. Following fertilization, ciliated planulae developed. Their mean size was 15015 lm (n=100). Eight hours after spawning, planulae were observed to contain zooxanthellae. Although spawning was not observed in the ®eld, after the spawning date no gametes were found in histological sections of coral samples collected at sea or from colonies kept in aquaria. At all localities (western and eastern Mediterranean), the peak period of gonad development coincided with the highest water temperature in August, and spawning took place in September with the beginning of temperature decrease (Fig. 6). A signi®cant correlation between gonad development and seawater temperature was found in O. patagonica colonies at the Israeli and French sites: r2=0.91 (P
