Comparative trace element trends in Posidonia oceanica and its sediments along the Turkish-Mediterranean coast


 Concentrations of eight trace elements (V, Cr, Ni, Cu, Zn, As, Cd, and Pb) were investigated in Posidonia oceanica blades and in the sediments sampled from 23 stations along the Turkish-Mediterranean Sea coast. Trace element (TE) concentration in both sediment and P. oceanica showed a statistically significant distribution among sampling stations. Most of the TE concentration in samples varied remarkably among stations without a consistent trend. Concentrations of Zn, Cd, Cu, and Pb in sediment of all of the sampling stations were determined as non-polluted based on the comparison with the sediment quality guideline index. More than quarter of the sampling stations were found as moderately or heavily polluted for Cr, Ni, and As. The mean TE concentrations found in the sediment sample in the present study were similar to the concentrations reported from the different parts of the eastern Mediterranean Sea. TE concentrations in P. oceanica were generally lower compared to the concentrations in the sediment except for Cd, Zn, and Cu. While positive correlation was found for Ni concentrations between sediment and P. oceanica samples, negative correlation was detected for V, Cr, Zn, Cu, As, and Cd concentrations between sediment and P. oceanica. The highest bioaccumulation factor in P. oceanica was calculated for Cd. The study area of the present study, especially the western sites (provinces of Antalya and Muğla) hosts millions of tourists annually and under the influence of intensive human activities in summer. Thus, coastal waters are heavily exposed to TEs and significantly positive correlations were detected between the anthropogenic TE pollutants (As, Cd, Cu, Zn, Pb, and V) and natural sourced TE (Ni and Cr).


Introduction
Posidonia oceanica (L.) Delile is an endemic owering seagrass species of the Mediterranean Sea inhabiting coastal waters between 0-40 m depth. They attach to the sedimentary seabed with their roots, from which they take up nutrients, and forms a habitat called meadow which extends over wide areas. These large underwater meadows are an important part of the ecosystem providing habitat for breeding, shelter, feeding, shore protection, nutrient cycling, and several other bene ts (Barbier et al. 2011;Short et al. 2011;Romero et al. 2015). Posidonia oceanica is sensitive to environmental deteriorations caused by human activities (Ruiz and Romero 2003;Holmer et al. 2008). The local decline in P. oceanica meadows has been reported from different parts of the Mediterranean sea due to anthropogenic disturbances (Montefalcone et al. 2009;Guillén et al. 2013).
Trace element contamination in the marine ecosystem is major concern due to their environmental persistence, biogeochemical recycling, bioaccumulation, and potential ecological risks. The Mediterranean Sea coast of Turkey is densely populated and most of the towns discharge their wastewater directly to the sea. Industries and agriculture in major cities of the region such as Muğla, Antalya, and Mersin also lead to trace element contamination through waste discharges. Tourism on the other hand also cause the similar form of pollution as any other industry. The Mediterranean Sea coast of Turkey is one of the leading tourist destinations that welcomes millions of tourists every year in summer and tourism is mainly concentrated in the coastal areas. Most of the coastline is urbanized and their population multiply during the holiday seasons and so the adverse effects on the marine environment multiplies.
Posidonia oceanica is suggested as an ideal biological indicator for biomonitoring (Debono and Borg 2006;Romero et al. 2015;El Zrelli et al. 2017) because of their wide distribution across the Mediterranean Sea, sensitivity to the environmental changes, abundance, and easy collection. (Pergent-Martini et al. 2005). Previous studies have shown the ability of P. oceanica up taking trace elements (TE) by its roots and leaves from the marine environment (Lafabrie et al. 2009) and its ability to re ect the former and current presence of the TEs in the environment (Pergent-Martini 1998). Posidonia oceanica has been extensively used as a bio-indicator for TE pollution (Pergent-Martini 1998;Lafabrie et al. 2007;Zhou et al. 2008;Zakhama-Sraieb et al. 2014). Perennial tissues of the species store TEs for long-term periods (Romeo et al. 1995;Lafabrie et al. 2009) and re ects the environmental status of TEs (Tovar-Sanchez et al. 2010). Sediment on the other hand, serves as a sink and source of trace metals. The accumulation of heavy metals in sediment affects the benthic organisms (Huang et al. 2020).
To best of our knowledge there is only a single study heavy metal contamination in the Turkish Mediterranean coast which is a local study performend in beach sediment in Mersin, Turkey (Yalcin 2009). A study is lacking to associate the trace element contamination in sediment with bioindicator species in the region. In the present study, we collected sediment and P. oceanica samples in order to de ne the current risk status of TE concentrations both in the sediment and P. oceanica throughout the Mediterranean Sea coast of Turkey. The main objectives of this study were i) to compare TE concentrations between the sediments and P. oceanica and ii) to compare TE status across the Mediterranean coast of Turkey.

Material And Method Sample Collection and Preparation
Sample collection was performed in 2019 (June-July) during acoustic eld survey for estimation of P. oceanica biomass. Posidonia oceanica and surface sediments were collected from 23 sites along the Mediterranean Sea coast of Turkey ( Figure 1). P. oceanica samples were collected by scuba diving between 10-40 meters and sediment samples were collected by a van Veen grab. Samples were placed in the sterile zip-locked bags and transported to the laboratory without breaking the cold chain and stored at -80°C until analyses.
In the laboratory, leaf blades (the highest accumulation has been reported in this part based on the works of Lafabrie et al. 2008;Zakhama-Sraieb et al. 2016) of the P. oceanica were separated, rinsed with ultrapure water to remove exogenous substances, and dried at 45 ºC for 24 hours. Dried blades were powdered in a mortar. Sediment samples (100-150 g) were placed in the glass petri dishes and dried at 4 ºC until constant weight. Dried sediment was homogenized in a mortar and then sieved through a 2 mm sieve. Each sample was digested using HNO 3 (Suprapur, Merck) and H 2 O 2 (Suprapur, Merck) according to US EPA 3050B method (EPA 1999). Brie y, 0.3 g of samples (dried sediment and blades) were placed into the extraction tubes containing 5 ml of HNO 3 (65% Suprapur, Merck) and heated on the block heater at 95±3ºC for 2.5 hours. After cooling, 2.5 ml of H 2 O 2 (30% Suprapur, Merck) was added, and tubes were heated on the block heater at 95±3ºC for another 2 hours. Next, the tube caps were removed, and samples were evaporated until having 2 ml of the mixture (leftover). Digested samples were diluted with ultrapure water and ltered through PTFE lter (0.45 µm pore size) and stored at 4ºC until analyses.

Trace Element Analysis
Concentrations of Vanadium (V), Chromium (Cr), Nickel (Ni), Copper (Cu), Zinc (Zn), Arsenic (As), Cadmium (Cd), and Lead (Pb) in the sediment and P. oceanica blades were measured by inductively coupled plasma mass spectrometry (ICP-MS) (Agilent, 7800). The accuracy of the analytical method was checked by using certi ed reference materials of estuarine sediment (BCR 277R) and seaweed (ERM-CD200) ( Table 1). Analytical measurements were performed in triplicate. Internal standards were used for quality control and quality assurance. The measured values were the mean±standard error (SE) and expressed as mg kg -1 (dry weight).

Statistical Analysis
The normality of the variables was tested using Shapiro-Wilk test. Homogeneity of variance was checked with the Levene test. Differences between the metal concentrations among stations were evaluated by one-way analysis of variance (ANOVA). Signi cantly different concentrations were determined by a post hoc test of Tukey HSD. Pearson correlation analysis was implemented to test for signi cant relationships among measured TEs. Parametric (Pearson) correlation coe cient was calculated to evaluate the relationship between the TE concentrations in the sediment and P. oceanica. Factor analysis (FA) was further implemented to to assess the major sources of the TEs in the sediment samples. The ability of P. oceanica to accumulate the TEs that present in the sediment (bioaccumulation factor) was also calculated with the following formula: Where C blades is the TE concentration in the P. oceanica blades and C sediment is the TE concentration in the sediment.
The data analyses were conducted by using R (ver. 4.0.3) (R Core Team 2020) and the visualizations were prepared by using ggplot2 (Wickham 2016) and cowplot (Wilke 2020) packages.

Results And Discussion
Trace elements in Sediment Sediment is vital, particularly for the coastal ecosystem, yet is under pressure from several anthropogenic pollutants due to the increasing population in the coastal regions. TEs, one of the most important pollutants, will go under cyclic sedimentation and accumulate in the surface sediment after a certain period following entering the coastal  (Table S1). The distribution of TEs in the sediments sampled from different locations was given in Fig. 2. Although most of the TE concentrations in the sediments varied signi cantly (p < 0.05, ANOVA, post-hoc: Tukey) among stations, not a consistent trend was observed in these variations (Fig. 3, Table S1). TE concentrations in the sediments were listed as follows: Ni > Cr > V > Zn > As > Cu > Pb > Cd. Factor analysis was used to assess the major sources of the TEs in the sediments collected along the Mediterranean Sea coast of Turkey. Two components (eigenvalues values > 1) were extracted (  (Salomons and Förstner 1988). Additionally, Amorosi (2012) determined that Ni and Cr, are caused by terrestrial transport especially in the coastal sediments, and it even has been reported that Ni and Cr go above the threshold limits with these transfers (Amorosi and Sammartino 2007).
Pearson's correlation test was applied to determine the relationship between the TEs detected in the sediments and the data were given in Table 3. There was a statistically signi cant (p < 0.01) relationship between Ni and Cr which are thought to have a similar contamination source (r = 0.88) based on the FA results (second component) ( Table 2).
On the other hand, V, Cu, Zn, As, Cd, and Pb, which are thought to have anthropogenic contamination based on the FA data ( rst component) given in Table 2, were found to have signi cant relationships among the listed TEs (Table 3). The researchers try to evaluate the sediment quality by comparing the TE values obtained from sampling with the indexes speci ed by various legal organizations. One of these indexes is the Sediment Quality Guideline (SQG) speci ed by the US EPA. According to the SQG sediments are classi ed under 3 categories as non-polluted, moderately polluted, and heavily polluted (Perin et al. 1997). TE concentration values obtained in this study were compared with the SQG values (Table S1). Zn, Cd, Cu, and Pb were determined as non-polluted TEs in all stations. Cr was found non-polluted in 39% of the stations, moderately polluted in 39%, and heavily polluted in the remaining stations while Ni was found to be non-polluted in 48%, heavily polluted in 39%, and moderately polluted in the remaining stations. For As, 65% of the stations were heavily polluted, while 17% of the remaining stations were nonpolluted and the other half of the stations (17%) was moderately polluted (Table S1). were generally similar to the concentrations reported in the other areas (Table 4). Indeed, in a study conducted in the coastal area of Morocco (El-Sorogy and Attiah 2015) the Ni, Zn, As, and Pb values were obtained higher than the values determined in our study (Table S1 and Fig. 3). The values were found to be similar or lower than the literature in terms of Pb data obtained in our work (Roussiez et al. 2005   Trace elements are divided into two groups as essential (As, Cr, Cu, Ni, V, and Zn) and non-essential (Cd and Pb) elements in the biological system (Tacon 1987;WHO 2004). Besides, non-essentials can be toxic even at very low concentrations, while essential ones can only be toxic to marine life above a certain concentration (EPA 1999;Nordberg et al. 2007). Also, these TEs are among the elements frequently monitored by the researchers in the . Thus, these elements were preferred in our study and their accumulation in the P. oceanica was examined. The distribution of the TE concentrations in P. oceanica sampled along the Mediterranean Sea coast of Turkey was presented in Fig. 2. With a widely varying range, minimum and maximum TE concentrations were detected as follows (mg kg − 1 ): As, 0.01-5.07; Cd, 0.03-3.07; Cr,Cu,Ni,Pb,V,Zn,. On the other hand, TE concentrations in blades of P. oceanica were as follows: Zn > V > Cr > Ni > Cu > Pb > As > Cd. Trace element concentrations in the blades showed a statistically signi cant distribution among sampling stations ( Fig. 3 and Table S2). Although most of the TE concentrations in P. oceanica blades varied remarkably among stations, no consistent trend was observed in these variations. Lafabrie et al. (2007) stated that the source of these elements in the coastal meadows is mostly anthropogenic and originating from the coastal settlements and heavy industrialization or harbors and rivers. The region in which the study was conducted, especially the provinces of Antalya and Muğla, stands out for their summer tourism and these provinces both host nearly 20 million tourists annually. Furthermore, these areas are under the in uence of intensive human activities such as hotel stayings, beaches, private travels, and sightseeing boat tra cs (Mutlu et al. 2020). Due to these activities, it is mainly considered that the coastal waters are heavily exposed to TEs, especially owing to untreated wastewater and excessive sea tra c resulting from these activities along the coastal line. Therefore, the FA results overlap the data showing the TEs mostly derived from the anthropogenic activities except for Ni and Cr (Table 2). Besides, when the correlation of the TEs was examined with each other, a signi cant positive correlation was detected between the elements that are determined to be anthropogenic (As, Cd, Cu, Zn, Pb, V) and naturally sourced (Ni and Cr) ( Table 3).
When we compare the trace element concentrations detected in the sampled sediments and P. oceanica TE concentrations in the blades were mostly lower compared to the concentrations in the sediments except for Zn, Cu, and Cd. A good bio-indicator is expected to have a good correlation between its TE concentration of the surrounding water column or sediment and the concentration in its tissues (Bonanno and Orlando-Bonaca 2018). In accordance, several other studies have revealed that P. oceanica re ects the TE status in the environment (Lafabrie et al. 2007;Bravo et al. 2016). Therefore, the relationship between the concentrations found in the sediments and the blades of P. oceanica were analyzed by linear regression analysis (Fig. 4). Different sampling locations showed a positive relationship ( Fig. 4; R 2 = 0.43) in terms of sediments and the P. oceanica for Ni concentrations, (this element was also found as the highest element in terms of average abundance detected in the sediment samples), however, a negative relation for the V, Cr, Zn, Cu, As, and Cd was recorded (Fig. 4). Besides, there was no relationship between Pb concentrations in the sediment samples and the P. oceanica (R 2 :0.0004) (Fig. 4). Furthermore, as in our study, Bonanno and Martino (2017) reported that Ni and Zn have a positive relationship with the sediments and P. oceanica. Bonanno and Martino (2017) also stated that there was no relationship between their sediments and P. oceanica in terms of Pb concentrations. On the contrary, some other researchers reported a positive relationship between the sediment and P. oceanica metal accumulations (Bravo et al. 2016;Bonanno et al. 2020 Bravo et al (2016), were in accordance with the results of our current study, only our Ni and Cd values were found to be higher from another study conducted by Bonanno and Martino (2017) Trace element concentrations in the P. oceanica reported in the other coastal areas of the Mediterranean Sea were listed in Table 5. Trace element concentrations in our study were mainly within the range of the values reported for P. oceanica in previous research (Warnau et al. 1995;Baroli et al. 2001;Lafabrie et al. 2008;Serrano et al. 2011;Ferrat et al. 2012;Di Leo et al. 2013;Zakhama-Sraieb et al. 2016;Saliha et al. 2016;Bonanno and Di Martino 2017;El Zrelli et al. 2017;Zeghdoudi et al. 2019). However, Cr and Pb were the only elements that concentrations were generally found higher compared to previously reported data (Table 4).

Conclusion
In the present study, trace element concentrations were evaluated in Posidonia oceanica and sediment collected from 23 sampling sites in the Turkish-Mediterranean Sea coast. Results have shown that some of the sampling stations are signi cantly contaminated by Cr, Ni, and As. Trace element concentrations in P. oceanica and in sediment were within the range of the values that reported from the different parts of Eastern Mediterranean Sea. It is worth mentioning that the P. oceanica blades show different patterns of metal contamination than sediment. The range of values for these trace elements provide useful information regarding the presence of anthropogenic contamination and possible sources.
Declarations Acknowledgement Sampling of the study was carried out during an acoustic survey study within the framework of the project funded by the Scienti c and Technological Research Council of Turkey (TUBITAK: 117Y133). Experimental procedures of the study were funded by the Scienti c Research Projects Coordination Unit of Karadeniz Technical University (Project No. FHD-2020-9100). The authors would like to thank the crew of the R/V "Akdeniz Su".
Ethics approval and consent to participate Not applicable

Consent for publication
Not applicable Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
RÇÖ and EM acquired the nancial support for the project leading to this publication. KG, RÇÖ, and AŞ contributed to the study conception and design. Sampling was performed by YÖ, AŞ and RÇÖ. KG performed trace element analysis. The rst draft of the manuscript was written by RÇÖ and KG. All authors read and approved the nal manuscript. Trace element concentrations (average ± standard error; dry weight, mg kg-1) in the sediments (yellow bars) and the Posidonia oceanica (purple bars) collected from 23 different stations along the Mediterranean Sea coast of Turkey.

Figure 4
The relationship between trace element concentration in sediment and Posidonia oceanica sampled along the Mediterranean Sea coast of Turkey. Each dot shows the result of different samples.