Seasonal dynamics of soft-bottom polychaetes in Izmir Bay (Aegean Sea, eastern Mediterranean)

Polychaetes comprise the majority of faunal components of soft-bottom benthic communities (Knox, 1977) and can be used as “markers” of different ecological conditions (Gambi and Giangrande, 1986). Some polychaete species that are very tolerant of environmental variables have also successfully been utilized in pollution monitoring studies (Reish, 1955; Bellan et al., 1988). Increase in industrialisation, human population and marine transportation in Izmir Bay and its periphery since the 1970’s has resulted in chronic pollution particularly in the inner part of the bay which has weak circulation. The main pollutants affecting the bay water quality are organic matter, suspended SCIENTIA MARINA 70S3 December 2006, 197-207, Barcelona (Spain) ISSN: 0214-8358

Due to eutrophication, the primary production (Chlorophyll-a) of the bay was estimated to be almost 6-8 times higher than the eastern Aegean production (Balci et al., 1995).The polluted zone of the bay has gradually extended from the inner region of the bay towards the relatively less disturbed middle and outer regions.Since the Gediz Plain is subjected to intensive farming and industrialization, the Gediz River, which enters the outer part of the Izmir Bay (Fig. 1), contributes considerably to the prevailing pollution in Izmir Bay (Ergen et al., 2002a).In early 2000, a wastewater treatment plant began to treat domestic and industrial wastes discharged into the bay.This plant treated the wastes at ca. 60% capacity between 2000 and 2001.A recent study showed that the capacity of the wastewater plant is sufficient for removal of nitrogen from the wastes, but inadequate for removal of phosphate (Kontas et al. 2004).
The present study was carried out before and after the installation of the waste water treatment plant, and aimed to determine the spatio-temporal distributions of soft-bottom polychaetes in Izmir Bay between 1997 and 2002.

The study area
Izmir Bay is one of the largest embayments along the eastern coast of the Aegean Sea.From the standpoint of its topographical and hydrographical characteristics, Izmir Bay was divided into three sections: inner bay, middle bay and outer bay.The inner bay, which extends between the head of the Bay and the Yenikale lighthouses and has a depth ranging from 0-20 m (average 10 m), is heavily polluted mainly by domestic and industrial activities.The middle and outer parts of the bay, with average depths of 16.3 m (range: 0-40 m) and 29 m (range: 0-70 m) respectively, receive less pollutants.In the outer bay, the Gediz River is the main source of pollution.Physico-chemical characteristics of the bay have been recently well documented: water and sediment quality by Kucuksezgin (1996), Aksu et al. (1998) and Atgin et al. (2000); hydrogen suphide and odour problem by Müezzinog ˇlu et al. (2000); physical features by Sayın (2003); nutrient and iron enrichment by Bizsel and Uslu (2000); eutrophication processes by Balci et al. (1995), Bizsel et al. (2001) and Kontas , et al. (2004).
According to Sayın (2003), Izmir Bay seawater is composed of five different water masses.These are the polluted inner Bay waters, the transient water SCI. MAR., 70S3, December 2006, 197-207 mass between clean waters and polluted waters, the clean water masses greatly influenced by a local upwelling near Gulbahce Bay, and the inflows of the Gediz River and Aegean Sea.Izmir Bay is under the influence of northerly winds throughout the year.
A recent seasonal study performed in Izmir Bay between 1996 and 2001 by Kontas , et al. (2004), who analysed water and sediment quality at stations where we collected the benthic material, showed that the temperature in the bay varied from ca. 10ºC in December (average value: ca.14ºC) to 28.6ºC in August (average value: ca.22ºC).The salinity varied from 21.4 (at a station near the Gediz River) to 39.9 psu (average value: ca.39 psu); the density from 16.01 to 30.90 kg m -3 (average value: ca.29 kg m -3 ).The Secchi disc depth, an indicator of relative primary production and pollution levels in the marine environment, had the maximum score (29 m) in the outer Bay in October and the minimum score (0.3 m) in the middle-inner parts of the bay in July.

Data collection and analysis
Cruises of R/V K. Piri Reis to Izmir Bay, within the framework of the project "Izmir Bay Marine Research Project" supported by Izmir Metropolitan Municipality, collected seasonal soft-bottom samples using a Van Veen grab sampling an area of 0.1 m -2 at 29 stations between 1997 and 2002, except for winter in 1999 and 2000.Only one sample was taken from each station during the study period.On the board, soft-bottom samples were sieved through 0.5 mm mesh and the retained material was placed in separate jars containing 4% seawater formaldehyde solution.In the laboratory, samples were rinsed in fresh water and sorted according to taxonomic groups using a stereomicroscope, and preserved in 70% ethanol.Afterwards polychaetes were identi-fied and counted using both stereo-and compound microscopes.The wet weight of polychaetes was determined by a sensitive balance.
In order to interpret the quantitative data, Shannon-Weaver's diversity index (H') and Pielou's evenness index (J') were applied to the abundance of the species.The pooled abundance data of species, obtained per sampling station in each season, were analysed using cluster and non-metric multidimensional scaling (nMDS) techniques, based on the Bray-Curtis similarity, using the PRIMER package (see Clarke and Warwick, 2001).SIMPER analysis was performed in order to identify the percentage contribution of each species to the overall similarity within each group that was assessed according to results of the cluster analysis.
Of the 44 families, Syllidae and Sabellidae had the highest number of species, accounting for 13.7% and 6% of the total number of species, respectively The results of nMDS and cluster analysis based on the pooled abundance data of the species collected between 1997 and 2002 are illustrated in Figure 2. According to these analyses, there are four distinct species associations (A-D) in the area, each having an average similarity higher than 50% (Table 1).However, group A has two stations (23 and 24) connected to each other at high similarity (65%) and one station (29) linked to these stations with relatively low similarity (39%), all being located in the polluted inner part of Izmir Bay.Although all these stations were dominated by opportunistic polychaete species such as Ophiodromus pallidus (max density: 720 ind.less population density of Sigambra tentaculata and no specimens of Lumbrineris spp., Cossura soyeri and Chaetozone spp., which were dominant components of the polychaete assemblage at stations 23 and 24.Streblospio gynobranchiata, which appeared in abundance on the polluted bottom of the inner part of the bay in 2002, is one of the seven species contributing greatly to the similarity score calculated among the stations 23, 24 and 29 (Table 1  Figure 3 indicates the mean numbers of species, individuals, diversity index, evenness index and biomass values estimated seasonally in each group from 1997-2002.Only the diversity of species present varied significantly in relation to seasons (1-way ANOVA, p<0.05).No significant difference was found in the number of individuals, diversity, evenness and biomass (p>0.05) between the seasons.
The lowest mean number of species was always found in group A (except in spring 1997) and the highest mean number of species generally in group C. Stations located in the polluted inner bay were represented by a very low number of species in summer and a relatively high number of species in winter and spring.Azoic conditions were recorded at station 29 in summer and sometimes in autumn from 1997-2001.However, such conditions did not occur in 2002, even in summer, and station 29 possessed at least 5 species (maximum 8) in all seasons of 2002.
Mean polychaete density varied among seasons and associations, but attained its highest scores in group A in spring, autumn and winter.As Capitella capitata capitata (6820 ind.m -2 ) and Polydora cornuta (1940 ind.m -2 ) exceptionally dominated station 29 in winter 2002, the mean density score had a peak in this period, with a high standard error value due to high polychaete density (9830 ind.m -2 ) at station 29 but relatively low densities at stations 23 and 24 (ca.1600 ind.m -2 ).Group D is characterized by rel-atively low polychaete densities [range: 60 (at station 3 in winter 1997) -2020 (at station 9 in autumn 1997) ind.m -2 ].The mean polychaete densities of groups B and C ranged from 190 (winter 1997)

DISCUSSION
Faunistic analysis of soft-bottom polychaetes inhabiting Izmir Bay showed that the area is biodiverse, harbouring a total of 396 species, of which 32 are new records for the coasts of Turkey.The previous studies carried out in the bay also pointed out its high biological diversity (Geldiay and Kocatas ,, 1972;Ergen, 1976;Kocatas ,, 1978;Çinar et al. 1998).However, domestic and industrial pollution, which was first detected in the bay in the beginning of the1970's, resulted in a decrease in the living components of the ecosystem of the inner part of the bay and favoured dense settlement of opportunistic polychaete species in the area (Geldiay and Ergen, 1972;Ergen, 1979;Geldiay et al. 1979;Kocatas , et al., 1986;Koçak et al., 1999).
Eutrophication is frequently reported particularly from the inner part of the Bay (Balci et al., 1995).The concentration of total inorganic nitrogen was 3.8 and 668 μM in the outer and inner parts of the bay, respectively (Kucuksezgin, 1996).The bay is also very productive; the average primary production values were calculated as 264, 463 and 726 mg C/m 2 /day in the western, central and eastern parts of the bay respectively (Balci et al., 1995).Azoic conditions and odorous biogenic sulphurous gases were reported to be serious problems in the area (Kocatas ,, 1980;Müezzinog ˇlu et al., 2000;Dog ˇan et al., 2005).Kocatas , et al. (1987) carried out a long-term monitoring study in the inner part of the bay between 1974 and 1983, and did not find azoic conditions even at the station located in the innermost part of the bay.They reported that the station had an average of 16 benthic species in 1974, whereas it possessed only 3 species in 1983.After the mid 1980's H 2 S odour problems, created by organic pollution and anoxia in the shallowest and most stagnant part of Izmir Bay continuously fed by polluted inputs of the rivers, were evident, especially during the long warm and dry season.However, an improvement in the environmental quality of the bay was expected after a new waste water treatment plant came into operation in early 2000.The present study includes periods before and after the establishment of the treatment plant and may indicate the efficiency of the plant in controlling and reducing industrial and domestic wastes pouring into the bay.As can be seen from Figure 4, azoic conditions at station 29, which is located in the innermost part of the bay, were found in every summer (except for 2002), in autumn 1997 and 1998 and in spring 1998.However, benthic samples taken during 2002 at this station had at least 5 species (autumn).The most interesting finding is that the sample taken in July 2002 at station 29 contained six species: Sigambra tentaculata (20 ind.m -2 ), Glycera unicornis Savigny, 1818 (40 ind.m -2 ), Polydora cornuta (30 ind.m -2 ), Prionospio fallax (230 ind.m -2 ), Streblospio gynobranchiata (990 ind.m -2 ), and Capitella capitata capitata (30 ind.m - 2 ) This suggests that the plant is effective in reducing pollution and has made the area more favourable even in summer for the settlement of polychaetes.Polychaetes are known to be colonisers of bottoms where a pollution abatement programme had been established (Pearson and Rosenberg, 1978).However, Kontas et al. (2004) found no significant differences in the overall average values of nutrient concentrations between the samples before and after the plant became operational.
The species composition of polychaetes encountered in the bay during the study is somewhat similar to those previously reported from the soft-bottom communities of the Mediterranean Sea.The most striking difference is that the present study reports the settlement of non-indigenous species such as Leonnates persicus (Wesenberg-Lund, 1949), Polydora cornuta and Streblospio gynobranchiata.The former is a Lessepsian migrant and has become established in Izmir Bay (Çinar et al., 2002).A recent study showed that the population density of this species inhabiting soft substratum of the northern part of the Levant Sea is positively correlated SCI. MAR., 70S3, December 2006, 197-207. ISSN: 0214-8358  with total carbonate and total inorganic carbon contents of the sediment (Ergev et al., 2003).Similarly, it was found at stations 11, 13, 27 and 28 in Izmir Bay, which receive high loads of organic material from the inner part of the bay and the Gediz River.This species formed a dense population (100 ind.m - 2 ) at station 13 in autumn 2002.The occurrence and impacts of the exotic spionid species, Streblospio gynobranchiata and Polydora cornuta, on the native fauna were documented in the inner part of Izmir Bay by Çinar et al. (2005).The other Lessepsian species found in the area are Lysidice collaris Grube, 1870, Metasychis gotoi (Izuka, 1902), Rhodine loveni Malmgren, 1865, Monticellina dorsobranchialis and Pista unibranchiata Day, 1963.However, their status of Lessepsian migrants should be re-analysed (Ergen et al., 2002b).
The results of multivariate analysis showed that there are four major species assemblages in the soft bottom of Izmir Bay (Fig. 2): the group developed in the polluted inner bay and characterized by relatively high population densities of opportunistic species such as Capitella capitata capitata, Malacoceros fuliginosus, Polydora cornuta and Ophiodromus pallidus; group B comprising stations not far from the sources of pollution discharges and under effects of inflows of the Gediz River; group C comprising shallow water stations, not strongly influenced by pollution and having coarse sandy mud; the last group D being located at the opening of the Bay, including deeper stations and having fine sandy mud.However, stations 16 and 22 within group B are linked to the other stations by relatively low similarity values.Station 16 has a sandy bottom, where Nephtys incisa, N. hombergi Savigny, 1818, Scoloplos armiger, Spio decoratus and Owenia fusiformis Chiaje, 1842, (typical species of sandy sediments and that can tolerate small amounts of mud and organic matter in the substrate (Gambi and Giangrande, 1986)), are represented by a relatively high number of individuals.The area around station 22 can be accepted as the transitional or semi-polluted zone as it comprised faunal elements from both adjacent communities.A total of 85 species were found in this station and the diversity index values ranged from 1.78 (summer 2002) to 3.59 (autumn 1997), with a mean value of 2.85 ± 0.12 SE.The relatively high number of species and diversity index values at this intermediate zone station can be explained as an "intensity effect":caused by the spatial complexity (mixed) of the sediment that results in a large number of niche spaces (Gambi and Giangrande, 1986) and the "edge effect" due to overlapping of adjacent communities (Dajoz, 1977).Station 14 within a Posidonia oceanica meadow had a specific polychaete assemblage characterized by high diversity and evenness, showing little affinity with unvegetated bottoms.Posidonia oceanica, a species endemic to the Mediterranean Sea, is known to host diversified polychaete assemblages both in its western (Gambi et al., 1995) and eastern (Çinar, 2003) regions due to its long, broad leaves and scaled rhizomes penetrating well into the sediment that provide a number of microhabitats and niches for both epifauna and endofauna.
According to statistical and faunisticanalyses, pollution, sediment structures and water depth seem to be the main factors governing the distribution of polychaetes in Izmir Bay.It is well known that depth, seasonality and particle size of sediment are the most important factors controlling distribution of polychaetes (Tom and Galil, 1991;Mackie et al., 1997;San Martín et al., 2000).Gambi and Giangrande (1986) and Mackie et al. (1997) found that depth affects both density and diversity of benthic communities, as the finer sediment increases with depth.The same result was also found in the present study where deeper stations located in the outer part of the bay had generally high diversity index values, with no conspicuous dominant species.Sardá et al. (1999) determined clear seasonal trends in the abundance and biomass of macroinfaunal assemblages at sublittoral stations of the Bay of Blanes, with peaks during spring, followed by a striking decline throughout summer and lower values during autumn and winter.The authors concluded that seasonal patterns of reproductive events are the main factors which explain such markedly repeated cycles.The seasonal dynamics of macroinfauna species was reported to be also influenced by the availability of food and the presence of predators (Sardá et al., 1995).However, Oug (2000) examined soft-bottom macrofauna from the fjords of northern Norway from 1979-1982 and reported that species number and densities varied irregularly with peak values occurring at different times of the year.He concluded that the lack of seasonality is mainly due to the stable physical conditions and the irregular nutrient input in the fjord basin.Apart from number of species that showed a significant difference between seasons, the seasonal fluctuations in number of individuals, and values of diversity, evenness and biomass are not statistically significant in this study.The same result was also found by Çinar et al. (1998), who investigated seasonal trends of faunal components associated with a Zostera marina bed in the Aegean Sea.
It can be concluded that pollution in the inner part of the Bay has been gradually diminishing since the waste water treatment plant became operational.Azoic conditions, which always occurred in summer in the inner part of the bay between 1997 and 2001, did not reoccur in 2002, suggesting that the area is in the process of recovery.Future monitoring studies to be undertaken in Izmir Bay will undoubtedly provide further information on the efficiency of the plant in reducing pollution in the area.
FIG. 1. -Map of the investigated area (Izmir Bay), with the locations of sampling sites.The station groupings (A-D) were made according to the cluster analysis presented in Figure 2.
FIG. 2. -Groups (A-D) derived from the Bray-Curtis similarity analysis.nMDS plot showing similarity grouping of stations.

TABLE 1 .
-Species contributing to similarity and dissimilarity within each groups (as shown in Figure2), with their average similarity.