Infralittoral ostracoda and benthic foraminifera of the Gulf of Pozzuoli (Tyrrhenian Sea, Italy)

The shallow water benthic foraminiferal and ostracod assemblages of the Gulf of Pozzuoli, located in the central Tyrrhenian Sea, were studied to investigate the relationship between calcareous meiofaunas and contaminant concentrations in bottom sediments exposed to prolonged industrial pollution. Both benthic foraminifers and ostracods displayed high-diversity and low-dominance, unusual features in highly contaminated environments. High-diversity values were possibly linked to the oligotrophic, well-oxygenated, and CaCO3-supersaturated coastal Mediterranean waters. The comparison with historical data suggested that assemblage composition changed in the last decades, with an increase in the relative abundance of benthic foraminiferal (Quinqueloculina seminulum, Bulimina elongata) and ostracod (Xestoleberis, Loxoconcha, Semicytherura rarecostata) taxa. They probably represent organisms tolerant to the environmental variations in the last decades. The relationships between granulometry and diversity indices, high correlation values between Quinqueloculina lata and heavy metal pollution, and the preference of the ostracod genera Urocythereis and Paracytheridea for very shallow marine waters were highlighted.


Introduction
Benthic foraminifers (Rhizaria) and ostracods (Crustacea) are meiofaunal groups generally provided with calcareous tests and valves, commonly preserved in sea bottom sediments. The composition of their assemblages reflects environmental conditions due to both natural and human causes. Anthropogenic activities produce various effects on shallow marine waters, including organic pollution, changes in sedimentation rates, increase in hydrocarbon and heavy metal concentrations, and eutrophication-induced hypoxia (Gooday et al., 2009;Yasuhara et al., 2012;Wilkinson et al., 2014) that, in turn, lead to an increase in the relative abundance of stress-tolerant foraminiferal species (Hayward et al., 2004;Alve et al., 2009;Frontalini and Coccioni, 2011;Ruiz et al., 2012) and frequently to ostracod diversity decrease (Alve, 1991;Mazzola et al., 1999;Cronin & Vann, 2003;Irizuki et al., 2018). The studies combining the analyses of benthic foraminifers and ostracods in areas where human-induced ecological variations occurred showed the high potential of calcareous meiofaunal assemblages as water quality indicators (Samir, 2000;Triantaphyllou et al., 2003Triantaphyllou et al., , 2005Vilela et al., 2003;Bergin et al., 2006;Pascual et al., 2008, Salvi et al., 2015.
In the present study, benthic foraminiferal and ostracod assemblages were studied from eleven samples collected in the infralittoral zone of the Gulf of Pozzuoli, a bay located in the Campania region (Southern Italy) with a narrow continental shelf, a shelf break at about 40 m bsl, a maximum depth of 110 m, and an average depth of ca. 60 m ( Fig. 1; Somma et al., 2016). The gulf is mainly exposed to winds and sea waves approaching from the southeast-southwest sector, with a maximum geographic fetch of 665 km for the 205°direction, 0.9-2.2 m average wave height, and a maximum wave height of 4.7 m in winter (De Pippo et al., 2008). The water circulation models (De Maio et al., 1985;Menna et al., 2008;de Ruggiero, 2016) of the gulf generally indicate two main flow patterns: (i) when the open sea currents flow toward the southeast, the inner waters of the bay are cut off in a slow cyclonic gyre; then, the coastal waters slow motion could favor turbidity and a high pollutant concentration; (ii) when the open sea currents flow toward the northwest, some branches enter into the bay; then, a fair renewal of sea waters occurs. Tides are negligible, with a syzygial tide amplitude of 0.35 m (Tammaro et al., 2021); therefore, the gulf is a wavedominated environment. Salinity, turbidity, and phytoplankton distribution are related to seasonal variation in the sea surface and column temperature, autumn-winter freshwater supply by rainfalls and land runoff, marine currents cell circulation: 37.1-38.6% salinity, 27-30°C sea surface temperature, and high phytoplankton biomass (Chl a  Somma et al., 2016 and the morphobathymetric and sedimentological surveys carried out for this research). The geographic coordinate system is WGS84 concentration [ 2 lg/L) were registered during the spring-summer season . The bay was exposed to prolonged anthropogenic disturbance, due to urban and industrial wastes, at least since 1885, when an armaments factory was built by the British company Armstrong Mitchell & Co. The eastern part of the gulf was under the influence of the Bagnoli steel plant from 1910 to 1990, to which the high levels of polycyclic aromatic hydrocarbons Ferrara et al., 2020), trace metals , and rare earth elements (Trifuoggi et al., 2018) in the sediment seem to be linked. Recently, some ecological and paleoecological investigations were performed on Recent (Balassone et al., 2016;Mangoni et al., 2016;Arienzo et al., 2020) and late Quaternary (i.e., from * 150 ka to historical times; Aiello et al., 2012;Amato et al., 2019;Petrosino et al., 2021) sediments of the Campania region coastal areas focused on benthic foraminiferal and ostracod assemblages. The present study aims to define the characteristics of the calcareous meiofaunal assemblages in the infralittoral zone (Peres & Picard, 1964;Peres, 1982) of an area showing high geoaccumulation values and to test a possible decrease in benthic faunal abundance and diversity in polluted bottom sediments. Our data, compared with the above-mentioned studies and previous investigations on Campanian infralittoral benthic foraminifers (Moncharmont Zei, 1964;Sgarrella & Moncharmont, 1993) and ostracods (Müller, 1894;Puri et al., 1964Puri et al., , 1969, may contribute to a more complete understanding of the relationship between meiofaunal assemblages and environmental parameters.

Material and methods
Eleven samples of very fine to coarse sands, and very fine gravels, were collected by a Van Veen grab above the shelf break of the Gulf of Pozzuoli (* 40 m bsl), in a water depth range between 7.5 m and 38 m, within the infralittoral zone. The grab collected superficial sediments, including the first ca. 5 cm of the seabed, related to texture and consistency of silt or sand deposits. Sharp & Nardi (1987) calculated, in this area, a sedimentation rate of about 4 mm/year and consequently the sampled sediments deposited, at most, in the last 15 years. The surface of the sampler is about 150 square cm (10 9 15 cm), while the volume of the sampled sea bottom surface sediment generally is about 750-1000 cubic cm. Samples were taken along one campaign in the spring of 2017 along transects and aboard a motor vessel. Bathymetry, grain size, number, and location of samples are reported in Fig. 1 and Table 1. For meiofaunal analyses, all the samples were oven-dried, and 100 g of dry sediment was taken. They were washed through 230-mesh (63 lm) and 120-mesh (125 lm) sieves, and the residues were oven-dried and examined with a reflected light microscope. A microsplitter was used (both MNI = minimum number of individuals per 100 g, and TNV = total number of valves per 100 g) assemblages. Q-mode cluster analysis (paired group as an algorithm, Rho as similarity measure) was performed to determine groups of samples with similar meiofaunal composition, using abundance values of all foraminiferal and ostracod (both MNI and TNV) species. Pearson's correlation coefficient was used to test for correlation between assemblage features, depth, major and trace elements, and polycyclic aromatic hydrocarbons of eight fine-grained samples; benthic foraminiferal and ostracod species with relative abundance greater than 5% in at least two samples were considered. The abiotic variables were subject to z-standardization. Analyses were carried out on the same set of samples used by Arienzo et al. (2017), Trifuoggi et al. (2017Trifuoggi et al. ( , 2018, and Ferrara et al. (2020) in their investigations on the distribution of polycyclic aromatic hydrocarbons (PAHs), trace metals (HMs), and rare earth elements (REEs). All the analytical determinations were performed in triplicate for each sample taken at each site. The quality of the analytical results is assured by participation in ring tests for the determination of HMs, PAHs, and REEs from sediments and similar matrices. Mean recoveries ranged from a minimum of 85% to a maximum of 97%. Grain size analyses were performed following the standard methodology of Folk and Ward (1957). Full methodological details on sampling techniques, geochemical and grain size analyses were reported in Arienzo et al. (2017), Trifuoggi et al. (2017Trifuoggi et al. ( , 2018, and Ferrara et al. (2020). Computation of diversity indices and statistical analysis were performed with STATISTICA 5 (Stat-Soft Inc., Tulsa, OK, USA). Six benthic foraminiferal species were present in all the samples, that is, Ammonia aberdoveyensis, Buccella granulata, Cibicides lobatulus, Elphidium crispum, Quinqueloculina seminulum and Triloculina schreiberiana. Assemblages were characterized by the genera Quinqueloculina (19 species) and Elphidium (10 species). Cibicides lobatulus was the most abundant species, with a Medium Relative Abundance (MRA) of 12.8%, followed by Tretomphalus concinnus (MRA = 5.54%), Siphonaperta aspera (MRA = 4.30%), Elphidium crispum (MRA = 3.97%), Asterigerinata mamilla (MRA = 3.75%) and Q. seminulum (MRA = 3.53%).
The number of species (S) was between 17 and 82, with discrimination between the three coarse-grained samples (TP2-1 = coarse sand; TP3-1 = medium sand; TP5-1 = very fine gravel) and the remaining eight samples, all made of fine or very fine sands. The former displayed a S range from 17 to 48, with the mean value of 34.33; the eight fine-grained samples had S between 52 and 82 (mean value = 61.25).
The number of specimens (I) showed a wide range, from 69 to 105,984. The three coarse-grained samples displayed a mean value of 484, whereas a I mean value of 51,540 was recorded for the remaining samples.
Taxa richness (S) ranged from 4 to 12 in the coarsegrained samples and from 31 to 57 in the remaining ones. In the former samples, abundance (I) was between 9 and 27 (MNI) and between 18 and 40 (TNV); in the latter samples, the mean value of I was 2065.5 (MNI) and 7781 (TNV). Shannon index H' followed a similar trend: in TP2-1, TP3-1, TP5-1 mean H' (MNI) was 1.68 and medium H' (TNV) was 1.56; in the assemblages occurring in the fine-grained sediments, medium H' (MNI) was 3.35 and medium H' (TNV) was 3.10.
Statistics. The cluster analysis ( Fig. 4) revealed two clusters of samples, obtained at a similarity cut-off level of 0.45. Cluster B consists of the coarse-grained samples TP2-1, TP3-1, and TP5-1, with low diversity-low abundance assemblages; cluster A includes the remaining eight fine-grained (fine and very fine sands) samples, characterized by high-diversity-high abundance assemblages. The three sediment samples grouped in Cluster B, consisting of medium sand, coarse sand, and very fine gravel, showed low geochemical accumulation. Since both low meiofaunal abundance/diversity values and low pollutant concentrations are highly associated with grain size (v. Discussion section), a correlation analysis including all the samples would provide results strongly influenced by granulometry. Consequently, we opted for performing the Pearson's correlation coefficient analysis on the eight fine-grained samples included in Cluster A. Results of Pearson's correlation coefficient analysis using meiofaunal assemblages, depth, major and trace elements, total organic carbon, and polycyclic aromatic hydrocarbons (Table 9) are reported in Table 10. The foraminiferal species Cibicides lobatulus and Elphidium crispum are common in all the samples. The assemblages included in Cluster A are characterized by the foraminifers Tretomphalus concinnus, Asterigerinata mamilla, Triloculina trigonula, and Elphidium punctatum and by the ostracods Semicytherura rarecostata, Loxoconcha ovulata, L. rhomboidea, Aurila convexa, and Xestoleberis communis. In Cluster B, the foraminiferal species Siphonaperta aspera, Quinqueloculina seminulum, Elphidium macellum, and the ostracods Urocythereis margaritifera and Pontocythere turbida characterized the assemblages.Our results show that the anthropogenic impact in the infralittoral zone of the Gulf of Pozzuoli, recorded in the geochemical accumulation (v. Table 9), was not reflected by diversity indices of the calcareous meiofaunal assemblages. The present findings were compared with the results of a previous study by Moncharmont (1964), based on a sampling carried out in 1961 by Harbans S. Puri and the Stazione Zoologica Anton Dohrn, where the characteristic species of the infralittoral zone were Ammonia beccarii, A. mamilla, C. lobatulus and T. concinnus Table 3 continued Samples TP1-1 TP1-2 TP2-1 TP2-2 TP2-3 TP3-1 TP3-2 TP3-3 TP4-1 TP4-2 TP5-1  Table 4 Ostracod absolute abundance [I(MNI) = minimal number of individuals per 100 g of sediment]; j indicates juvenile specimens      (splitted by Moncharmont in T. concinnus and Rosalina globularis); conversely, Q. seminulum, S. aspera, Bulimina elongata were very rare. Our data suggested that Q. seminulum and B. elongata, stresstolerant species , and S. aspera may have increased their abundance in the last decades.
The 1961 sampling campaign also provided bottom sediments of the adjacent Gulf of Naples. Their ostracofaunas were analyzed by Puri et al. (1964). In the infralittoral zone, the ostracod assemblages were dominated by Aurila, Urocythereis, Carinocythereis, Costa and Pontocythere (Cushmanidea in Puri et al., 1964) species, Tenedocythere prava (= Quadracythere (?) prava plus «Cythereis» polygonata, the latter name including the juveniles, in Puri et al., 1964) and S. incongruens. In the present infralittoral ostracofauna of the Gulf of Pozzuoli U. margaritifera and P. turbida are very common, but overall the assemblages are not dominated by trachyleberid species, being characterized by the genera Loxoconcha and Xestoleberis and by the species S. rarecostata.
It can be hypothesized that industrial and urban contaminants caused a meiofaunal change, more apparent in ostracod than in benthic foraminifers. Conversely, the high-diversity and low-dominance values showed that in oligotrophic and well-oxygenated waters (e.g., Hyams-Kaphzan et al., 2009, in a coastal area under the influence of sewage sludge with organic matter input), the diversity indices were not negatively influenced by the high concentration of pollutants in bottom sediments.
The sensitivity of ostracod and benthic foraminifers was also displayed by statistical analysis. Noteworthy correlations between ostracod dominance and sediment contaminants (polycyclic aromatic hydrocarbons, rare earth elements) encourage the use of calcareous meiofaunal assemblages as environmental bioindicators. The analysis confirmed the results of previous investigations, such as the tolerance of Q. lata to high heavy metal concentrations (Romano et al., 2009) testified by high correlations (C 0.76) with Ni, Pb, and Zn. Quinqueloculina lata was considered a pollution-tolerant species, by Elshanawany et al. (2011; and by Romano et al. (2013) in environment under strong anthropogenic pressure. The study of Mangoni et al. (2016) showed a correlation of this species with high concentrations of inorganic nutrients.
Pearson's correlation coefficient analysis of benthic foraminifers revealed that T. concinnus and taxa richness (S) correlated with As; foraminiferal dominance with Fe and Zn; Q. lata correlated with Ni, Pb, Zn, TOC and anticorrelated with water depth.
Total PAHs and total PD PAHs (Priority Dangerous PAHs) correlated with P. retifera (MNI), A. convexa and D and showed an inverse correlation with J (TNV). Rare earth elements correlated with A. convexa and D (TNV).

Discussion
The study of shallow water samples of the Gulf of Pozzuoli allowed us to investigate the distribution of meiofaunal calcareous assemblages in an area subject  Table 9 Concentrations (%) of TOC = total organic carbon; concentrations (mg/kg) of heavy metals, REE ? Y = rare earth elements ? Y, LREE (light rare earth elements), HREE (heavy rare earth elements), total PAHs (polycyclic aromatic hydrocarbons) and total PD PAHs (PD: Priority Dangerous PAHs: naphthalene (NAP), anthracene (ANT), benzo(b)fluoranthene (BbF), benzo(k)fluoranthene (BkF), benzo(ghi)perylene (BgP), benzo(a)pyrene (BaP); Trifuoggi et al., 2017;Ferrara et al., 2020)     to long-term industrial and urban pollution. All the sediments were collected within the infralittoral zone where previous researches displayed high geoaccumulation levels, especially in the eastern part of the bay Trifuoggi et al., 2017Trifuoggi et al., , 2018. Anthropogenic and natural environmental pressure may influence both the meiofaunal features (abundance, diversity indices, dominance) and the taxonomic composition of foraminiferal and ostracod assemblages. In the sediments of the Gulf of Pozzuoli, their diversity was strongly related to the grain size of the bottom sediments, being low in the coarse-grained samples and high in fine and very fine sands. The relationship between granulometry and meiofaunal diversity in shallow marine waters was investigated by several researchers, who identified indicative trends. Pokorný (1978) stated, as a general rule, that coarse sediments (e.g., oolites and clean sands), can hold only a limited number of ostracod species, whereas on pelitic bottoms and mud-mixed sands the diversity of the assemblages increases. A number of studies corroborated this relationship, showing high ostracod diversity in fine-grained sands and low diversity in coarser sediments (Hazel, 1975;Aiello et al., 2006), higher diversity on silts or mud-mixed sands, and lower in clean sands (Puri 1966(Puri , 1971) and negative correlation with granulometry in Quaternary sandy successions . On the other hand, on muddy bottoms, ostracod diversity frequently diminishes (Benson & Maddocks, 1964;Hong, 2016); consequently, it was suggested that the relationship between ''ostracod diversity and particle size fractionation is not unimodal but rather hump shape'' (Hong, 2016). A similar trend was observed concerning benthic foraminifers. Investigations on assemblages collected on muddy bottom recorded low diversity, increasing in sandy sediments (Diz et al., 2004;Debenay et al., 2005); in large-grain sands and coarser bottom sediments a decrease was observed (Samir & El-Din, 2001;Temelkov, 2008;Delaine et al., 2015). It has to be noted that some studies reported apparently contradictory results, showing, for example, high foraminiferal diversity in sheltered areas with finegrained sediments and high ostracod diversity in more exposed coarser-grained sediments (Morley & Hayward, 2014).
The characteristic of the assemblages occurring in the study samples supported the link between  granulometry and calcareous meiofaunal diversity. Coarse-grained sediments were present in the samples TP3-1 (medium sand), TP2-1 (coarse sand), and TP5-1 (very fine gravel), where the ostracod dominance is high, whereas abundance and diversity are low; foraminiferal taxa richness and abundance are low. In the infralittoral zone of the bay, muddy sediments were virtually absent, and Pokorný's statement was confirmed by higher diversity and lower dominance displayed by ostracod, and, to a lesser extent, by benthic foraminiferal assemblages in fine and very fine sandy samples.
Chemical analyses revealed lower concentrations of pollutants in coarse-grained samples. Arsenic, copper, mercury, lead, zinc, total polycyclic aromatic hydrocarbons (pahs; in particular anthracene, benzo(a)anthracene, benzo(b)fluoranthene, and chrysene) and priority dangerous pahs were significantly higher in fine and very fine sands. The recognized inverse relationship between grain size and anthropogenic chemical contaminants (Horowitz, 1985(Horowitz, , 1991Herut & Sandler, 2006) and the abovementioned link between meiofaunal remains and granulometry, could erroneously suggest the preference of ostracod and benthic foraminiferal assemblages for polluted bottom sediments. Consequently, to achieve reliable results, we considered separately the fine sandy samples and the coarser-grained sediments.
Our point of view is that neither the contaminants accumulated in the bottom sediments nor the dead assemblages are representative of the ecological conditions at the time of sampling. Instead, they are indicative of the environmental history of the bay during the last years or decades. On the base of the present data, it is hypothesized that the continued anthropogenic disturbance, primarily due to industrial wastes, had not resulted in a decrease in ostracod and benthic foraminiferal diversity and had its effect by changing the taxonomic composition of the assemblages. In Table 11, the mean values of diversity indices of the infralittoral assemblages of Monte di Procida (Mangoni et al., 2016) and Falerno-Domitio (Balassone et al., 2016) were reported. Both ostracod and benthic foraminiferal assemblages of the Gulf of Pozzuoli showed higher diversity and lower dominance in comparison with the nearby areas. The expected response to contaminant input was the decrease in meiofaunal diversity, nonetheless, some investigations on benthic foraminifers reported inverse trends Fernandez, 2009 andLi et al., 2015, in metal-contaminated waters) and highly diversified assemblages in polluted waters (Romano et al., 2009 andChoi &, in metal-contaminated waters; Armynot du Châtelet et al., 2011, in metal-contaminated, rich organic carbon waters).  Barras et al. (2014) suggested that in oligotrophic environments the diversity indices are not appropriate to describe water quality. Moreover, the common presence of miliolids implies coastal waters supersaturated in calcium carbonate (Aiello et al., 2018, and references therein) where the presence of industrial wastes did not result in a pH lowering. Reports of high-diversity ostracod assemblages in stressed environments are very rare (Amato et al., 2019;Aiello et al., 2020) which hampers to perform a proper comparison with the present study.

Conclusions
A quantitative study of benthic foraminiferal and ostracod assemblages, integrated with chemical and sedimentological parameters, was carried out along the coast of the Gulf of Pozzuoli. A total of 11 samples were collected within the infralittoral zone, to investigate the relationship between meiofaunal calcareous remains and pollution indicators preserved in the shallow marine bottom sediments of the bay. Assemblages were characterized by species typical of infralittoral Mediterranean environments such as the benthic foraminifers A. aberdoveyensis, B. granulata, and C. lobatulus, and the ostracods U. margaritifera and P. turbida. The genera Elphidium and Quinqueloculina (Foraminifera), and Semicytherura and Xestoleberis (Ostracoda) were highly diversified.
Despite the continued anthropogenic disturbance, testified by high geoaccumulation levels, the shallow bottom sediments of the Gulf of Pozzuoli yielded high-diversity, low-dominance assemblages. It was here hypothesized that the oligotrophic, well-oxygenated, and supersaturated in CaCO 3 shallow Tyrrhenian waters may promote ''complex trophic relationship'' and ''full exploitation of ecological niches'' as stated by Holbourn et al. (2013) in their investigations on foraminiferal deep-water assemblages. The diversity was linked to the grain size of the bottom sediments, being higher on fine and very fine sands, and lower on coarser sediments.
On the other hand, the comparison with the assemblages collected in 1961 in the Gulf of Pozzuoli and the Gulf of Naples showed an increase in the abundance of taxa that are suggested to tolerate to geochemical pollution (Q. seminulum, B. elongata, Xestoleberis, Loxoconcha, S. rarecostata). Statistical analysis confirmed the tolerance of Q. lata to a high level of heavy metals contaminants and the preference of U. margaritifera and P. paulii for the shallow part of the infralittoral zone. are grateful to the University of Naples Federico II for supporting our research in the APC for open access.
Author contributions Giuseppe Aiello involved in conceptualization, writing-original draft preparation, investigation; Diana Barra took part in conceptualization, investigation, writing-review & editing; Roberta Parisi participated in formal analysis, investigation; Carlo Donadio took part in data curation, investigation; Michele Arienzo involved in formal analysis, investigation, resources; Luciano Ferrara took part in formal analysis, investigation, resources; Maria Toscanesi involved in investigation, resources; Marco Trifuoggi participated in investigation, resources.
Funding Open access funding provided by Università degli Studi di Napoli Federico II within the CRUI-CARE Agreement. No funding was received for conducting this study.
Data availability All data generated or analyzed during this study are included in this published article [and its supplementary information files].

Declaration
Conflicts of interest The authors have no financial or proprietary interests in any material discussed in this article.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.