Annual trend of fish assemblages associated with FADs in the southern Tyrrhenian Sea By F. Andaloro 1 , D. Campo 1 , L. Castriota 1 and M. Sinopoli 2 1 ICRAM, STS Palermo, Palermo; 2 IAMC-CNR, Laboratorio di Ecologia della Fascia Costiera, Castellammare del Golfo, TP, Italy Summary A study on fish assemblage associated with fish aggregating devices (FADs) in Sicily was carried out between January 2000 and January 2001. With a fortnightly periodicity, 156 experimental hauls were carried out by means of a purse seine in a FAD-containing marine section banned to commercial fishing. A total of 14 229 fish specimens belonging to six families and 10 species was found. These species were Balistes carolinensis, Caranx crysos, Naucrates ductor, Seriola dumer- ili, Seriola fasciata, Tachurus picturatus, Coryphaena hippurus, Schedophilus ovalis, Thunnus thynnus, Polyprion americanus and showed all young-of-the-year undergoing a rapid growth. The applied ordination technique highlighted the existence of four assemblage periods describing the annual trend. The results confirm that fish assemblages associated with FADs are related to season, following a fish colonization tied to natural recruitment. Comparison of the ecological indices across the four periods showed that the assemblages in the periods from summer to winter were more structured than those in spring. The quantity of individuals also showed a strong variation peaking in the summer period. The results of this study reveal that FADs represent a particular nursery area for the associated species that could influence their survival. Introduction The distribution of certain juvenile and adult fish species is influenced by floating objects and artificial reefs (Hilborn and Medley, 1989). High fish-densities have been observed to aggregate around drifting algae (Dooley, 1972; Kingsford and Choat, 1985), driftwood (Hunter and Mitchell, 1967), jellyfish (Mansueti, 1963; Janssen and Harbison, 1981), carcasses of large animals (Kojima, 1966) and man-made objects such as rafts (Gooding and Magnuson, 1967). For this reason flotsam and other fish aggregating devices (FADs) have been used by fishermen throughout history to improve pelagic fish yields (Kojima, 1966; Massutı´ and Vidal, 1997). In Sicily the use of traditional FADs, called ÔcannizziÕ, is linked exclusively to the presence of economically important pelagic fish species along the coast, such as the greater amberjack Seriola dumerili (Risso, 1810) and the dolphinfish Coryphaena hippurus Lin- naeus, 1758 (Potoschi and Sturiale, 1996; Bono et al., 1998). In the Sicilian area FADs are usually set at sea by fishermen in August. The commercial catch begins in September and ends in January (Morales-Nin et al., 2000). Traditional FADs consist of palm fronds connected to floating objects (plastic bottles or polystyrene bodies) to facilitate their position, anchored with a rope to large stones (Fig. 1) (D’Anna et al., 1999). With the growing importance of flotsam and FADs in commercial fisheries (Fonteneau et al., 2000), scientific inves- tigations on fish assemblages associated with FADs and the causes of their association have rapidly increased in recent years (Dempster and Taquet, 2004). In the Mediterranean Sea, the seasonal trend of the fish assemblages associated with FADs has been studied off the Balearic Islands in the period between May and February (Massutı´ and Ren˜ones, 1994; Deudero et al., 1999), and in the Gulf of Castellammare (southern Tyrrhenian Sea) between July and November (D’Anna et al., 1999). Despite these investigations, no studies have ever been carried out with such devices remaining in place, even distant from that of commer- cial fishery. This paper reports the results of an annual study of fish assemblages associated with FADs in Sicily, based on experimental fishing data. Materials and methods An experimental field of 170 FADs was placed in the southern Tyrrhenian sea between Capo D’Orlando and the Eolian Islands, between 38°13¢–38°20¢N and 14°40¢–14°45¢E (Fig. 2). The study site was 600–800 m deep. The ÔcannizziÕ were set at sea in December 1999, forming a grid with a 500 m distance apart. Each FAD consisted of three palm leaves (Phoenix canariensis), six 2.5-L plastic bottles, a 30– 40 kg stone to anchor at the sea bottom, all connected by a 5 mm polypropylene line (Fig. 1). To minimize the effect of fishing activities on FAD-associated fauna, commercial fish- ing activities were banned within the experimental field during the study period. The study was carried out from January 2000 to January 2001 with a fortnightly periodicity and obtaining 26 sampling dates. On each sampling date, a modified purse seine with a cod-end mesh size of 2 mm was used following Massutı´ et al. (1999), in order to sample the entire length spectrum of all fishes occurring beneath the FADs. The hauls were conducted during daytime on six randomly chosen FADs. Each FAD was sampled only once during the study period to ensure independence of replicates. Therefore, a total of 156 hauls was carried out, corresponding to 156 FADs sampled among the 170 available. Indeed, hauling by purse seine beneath a FAD means removing the entire fish assemblage under that FAD, conditioning results of further sampling beneath that FAD. The collected fishes were identified to the species level, counted (n) and measured (TL, total length in mm). J. Appl. Ichthyol. 23 (2007), 258–263 Ó 2007 The Authors Journal compilation Ó 2007 Blackwell Verlag, Berlin ISSN 0175–8659 Received: July 28, 2006 Accepted: December 30, 2006 doi: 10.1111/j.1439-0426.2007.00860.x U.S. Copyright Clearance Centre Code Statement: 0175–8659/2007/2303–0258$15.00/0 www.blackwell-synergy.com