Metal Pollution and Health–Ecological Risk Assessment in an Intensely Burdened Coastal Environment of Greece, the Saronikos Gulf: A 50-Year Critical Review

Abstract

Among Greece’s coastal areas, the Saronikos Gulf encounters the highest environmental challenges due to heavy metal contamination, caused by extensive urbanization and industrialization. In the present study, online databases were used to identify research articles focusing on the levels, patterns, and origins of the heavy metals on the gulf’s seafloor published from 1974 to 2024. Thirty-three scientific papers were chosen to review the status of heavy metal pollution, set background values, and summarize the analytical methods used. Additionally, fourteen of them were used for a meta-analysis review. Geographic Information System (GIS) techniques were employed to map the sampling locations and heavy metal distribution per decade of the collected data, while the ecological status of the area was estimated via the application of indices such as the Pollution Loading Index (PLI), Potential and Ecological Risk Index (PERI), and Human Health Risk Assessment (HHRA) to the previously collected data. The review revealed that the Saronikos Gulf has mostly been studied in specific regions due to existent point sources. Additionally, the reassessment of the data referenced in the literature permitted integrative comparisons that could improve the management and sustainable development of the Saronikos Gulf.

1. Introduction

The interfaces between land and ocean—coastal areas—are of significant ecological value, as they support a wide range of habitats, biological diversity, fishing grounds, climate regulation, and carbon sequestration [1,2,3]. Despite and/or because of their environmental significance, coastal areas undergo intense industrial, commercial, and touristic uses [1]. They also serve as sites for urbanization and port facility growth, which presented increasing trends in the 20th century [2,4], resulting in coastal pollution through the deposition of various human contaminants in the aquatic environment. Thus, the effective management of coastal areas should be achieved by proper and continuous monitoring.

The sediment quality is considered a pollution estimate proxy, providing information on the spatial and temporal distribution of contaminants and recording the pollution evolution over time. In particular, the fine fraction of marine sediments serves as both a final repository and a source of human contaminants due to their co-precipitation with oxides, cation exchange, adsorption on particle surfaces, and assimilation by organic matter [5,6,7].

Strong indicators among contaminants that denote pollution levels are the metal and heavy metal concentrations [8]. They are used in a variety of industrial, domestic, and agricultural processes, while their production and usage keep increasing [3]. Although most metals in certain concentrations play an essential role in organisms’ development, elevated levels and non-essential heavy metals are toxic to flora and fauna [3]. Elevated concentrations of heavy metals can be attributed to human and/or lithogenic sources. Anthropogenic metals are reported in several studies as more bioavailable compared to those of lithogenic origin [2,9,10]. Due to their features of bioaccumulation, ecological toxicity, and non-biodegradation [11,12], which reflect risks to biota and human health, metals have therefore been consistently addressed as among the most dangerous contaminants [13,14,15]. In addition, heavy metals in sediments may also re-enter the water column through resuspension and cause secondary contamination, leading to possible environmental effects and threats to human health [16,17,18,19,20].

Considering the above, the heavy metal burden in Greece’s coastal sediments emerged in environmental studies in the early 1970s, conducted in the Saronikos Gulf [21,22,23,24,25]. The Saronikos Gulf hosts the capital of Greece and its largest port, whose operation has remained undiminished for the last 2500 years. Fifty years since the first published research on Saronikos sediment pollution due to heavy metals, published data on and knowledge of the coastal pollution in the area have been growing rapidly [26,27,28,29,30,31,32]. As the number of studies on marine sediment pollution escalates, review papers become a necessity to make researchers’ work easier and decision-makers more effective regarding the available data claimed over the years [1,3,7,33,34,35]. Literature reviews attempt to find fresh perspectives on earlier studies, point out any gaps in the literature, resolve contradictions between earlier research that appear to disagree, and identify areas of prior knowledge to avoid repeating work. Moreover, a systematic review compiles all the relevant studies on a particular topic and also assesses and examines their outcomes [3,36]. Additionally, meta-analysis is widely accepted as the preferred method to synthesize research findings in various disciplines. A systematic and meta-analysis review of the literature on the heavy metal pollution in the seafloor sediments of the Saronikos Gulf is therefore necessary given the fast-growing body of knowledge in this area. This will allow for the quantification of research efforts and trends, as well as for the identification of gaps for which further study is needed [3].

Various pollution indices, such as the Micropollutant Index, Geoaccumulation Index, and Enrichment Factor [21,37,38], were developed for the comprehensive evaluation of the contamination degree (e.g., Abrahim and Parker; Radomirović et al.; Kanellopoulos et al. [7,39,40]) and were used in the previous studies from the Saronikos Gulf [21,22,26,27,28,30,31,32,41,42,43,44,45,46]. In the present study, the Pollution Loading Index (PLI), Potential Ecological Risk Index (PERI), and Human Health Risk Assessment (HHRA) proxies were used since health risk assessment is an efficient method for estimating the dangers that heavy metals pose to human health through various exposure pathways [18,20,35]. Since the ecological and human risks of heavy metals in sediments have only recently been documented in the literature, this study investigates the ecological and human health risks of sediment-associated metals in large-scale coastal areas [20,47].

Thus, the current work provides a systematic review of studies on heavy metal pollution in the Saronikos Gulf over the last fifty years, between 1974 and 2024. Particularly, the following issues were addressed: (a) the consistency of the research trends in this field over the last fifty years, (b) the most and least studied topics and emerging trends regarding heavy metals and their anthropogenic sources, (c) the coverage of sampling sites over the last fifty years, along with the concentration and distribution of specific heavy metals in the seafloor sediments of the Saronikos Gulf using GIS mapping techniques, and (d) the magnitude of the environmental change due to human activities and the background values in the area. This methodological estimation approach should be useful, since no other previous systematic and meta-analysis review study in the literature has been conducted in the Saronikos Gulf summarizing the pollution history of the area and critically analyzing the past research methodologies, analytical instruments used, background levels estimated, dating techniques employed, and ecological assessments conducted. Additionally, the data on heavy metal concentrations estimated in the literature were used to visualize the pollution history and investigate the ecological and human health risks with the most novel approaches, thereby making this study applicable to other polluted areas with long pollution histories worldwide. Based on the results of the present critical review, measures and recommendations are proposed to cover the gaps in the previous works. Besides offering helpful information for efficient coastal management and environmental policy, this research may also propose baseline levels for future monitoring initiatives, thereby assisting in the preservation of local ecosystems.

2. Materials and Methods

2.1. Study Area

2.1.1. Geographical Setting

The Saronikos Gulf (Greece) is situated between the peninsulas of Attica and Argolis, adjoining the city of Athens in the Aegean Sea, and lies in the Eastern Mediterranean Sea, where the impact of pollutants is being intensified compared to that in open seas (Figure 1b,c). According to Robledo-Ardila et al. [48], wastewater discharge points co-exist with rapid urban development and increased tourism in coastal areas, resulting in high concentrations and a high variety of heavy metals (mainly Cu, Hg, Pb, Zn, Ni, As, Cr, Cd, Ba, and V) in the Mediterranean (Figure 1).

In Figure 1c, the wastewater discharging points are presented in the Mediterranean Sea, indicating the Saronikos Gulf, with high wastewater discharges among them. The gulf’s maximum depth is 416 m, southwest of the gulf (west of the Methana volcano). Saronikos is divided into two basins: an eastern one and a western one. The boundaries of the western basin are delineated by the imaginary lines connecting Methana–Aegina and Aegina–Salamina (Figure 1a). The eastern basin is divided into an inner one and an outer one. Elefsina Bay lies to the north and is connected to the rest of the gulf via two narrow and shallow straits (an eastern one and a western one). The gulf is characterized by several islands, and among them are Salamina and Aegina and smaller ones such as Psyttalia (Figure 1a).

2.1.2. Geology

The surrounding geological formations, draining into the Saronikos Gulf, comprise Plio-Quaternary deposits, Alpine limestones, schists, marbles, and ultra-basic ophiolite complexes [49]. In the eastern part of the gulf, marbles and crystalline limestones with occurrences of Jurassic schistoceratolithic facies are emplaced. Elefsina Bay (like Salamina Island) mostly comprises Triassic–Jurassic limestones, dolomites, and Upper Cretaceous limestones. The Saronikos Gulf forms the northwestern end of the South Aegean Volcanic Arc consisting of the volcanoes of Sousaki, Methana, and Aegina [7,50,51,52].

2.1.3. Water Circulation

Since there are no major rivers draining into the gulf (except occasional ephemeral streams discharging during short winter periods of heavy rainfall), the Saronikos Gulf is characterized by low freshwater inflows. In addition, strong seasonal density stratification, high nutrient accumulation, and poor environmental status portray the gulf’s hydrological conditions [23,53,54]. There are two prevailing surface circulation trends: cyclonic circulation resulting from northerly winds and anticyclonic circulation, which is less frequent and is due to southerly winds [55]. Winds and changes in barometric pressure contribute to a tide with a maximum range of 50 to 60 cm. Circulation in the upper Saronikos is predominantly wind-driven and therefore varies depending upon the prevailing seasonal wind direction [23]. Therefore, the dispersion of pollutants further into the Aegean Sea is low, given that the Saronikos Gulf is a semi-enclosed gulf with limited water exchange with the open sea.

2.1.4. Anthropogenic Pressures

The Saronikos Gulf is a highly human-impacted area, as it neighbors a highly urbanized and industrialized landscape (Attica, Elefsina, and the city of Athens) (Figure 1 and Figure 2). Elefsina and Keratsini Bays have been known since the early 1970s to receive water from the industrial, shipping, and domestic waste discharged from the Athens and Piraeus urban–industrial areas [23]. Environmental pressures can also be attributed to Piraeus Port, a Fertilizer Plant, shipyards and the ship repairing zone, the Naval Station, sunken, abandoned ships, and the former operation of two central sewers: an earlier one (in Faleron Bay) operating until the 1950s and a second one in Keratsini (the Athens Sewage Outfall) operating until 1995. The high organic mud content covering the natural seafloor and displaying an H₂S odor in Faleron Bay is attributed to the former operation (until the 1950s) of Faleron Central Sewer [25]. Additionally, 28 large industries discharging effluents of unknown quality and quantity to Elefsina Bay are enumerated by Griggs et al. [23] (Figure 2) (Table S1_1 Supplementary File_S1). In 1995, the Athens Sewage Outfall was replaced by the primary Wastewater Treatment Plant (WTTP) of Athens situated on the island of Psyttalia. About a decade later (in 2004), WTTP started operating as a secondary wastewater treatment plant [56].

Saronikos Gulf Monitoring.

The quality of the seafloor sediments was first studied in the early 1970s as a part of the Saronikos Systems Project (SSP) investigating the upper sediments of the gulf to assess the effect of the wastewater input from the Attica area [25]. Subsequently, about 10 to 15 years later, in 1986, the marine environment of the Saronikos Gulf was monitored in the framework of the National Monitoring Program for the Assessment and Control of Marine Pollution in the Mediterranean (MED-POL) (MAP/UNEP) under the supervision of the Greek Ministry of Environment, in a grid of 15 sampling stations [56]. Additionally, in 2011, the Saronikos Gulf became a member of the National Water Quality and Quantity Monitoring Network (Government Gazette 2017/2011) in the frameworks of the WFD (Water Framework Directive), and the beginning of systematic sampling was established in 2012 [57]. The Fertilizer Plant (FP), the Athens Sewage Outfall (ASO), Piraeus Port, and Elefsina Bay are the main pollutant factors of the Saronikos Gulf (Figure 2, Figure 3 and Figure 4).

Fertilizer Plant.

The Fertilizer Plant (FP) located in Keratsini Bay (Figure 2) was considered one of the main pollutant factors in the area. It has been reported since the 1970s that the FP discharges a red effluent discoloring the waters, while samples derived from red mud sediments near the FP were found holding elevated concentrations of As and Sb [22,23,42]. The area affected (by high metal concentrations) was estimated to extend to approximately 40 km². Elevated concentrations of not only V [46] but also Co, Fe, and Zn [58] were reported in sediments derived in front of the FP. The possibility of the FP effluents being transported to the northwest following the prevailing currents was also referred to, so that the pyrite-rich waste discharged by the FP has impacted the sediments around its outfall, extending their influence on the seafloor from the entrance of Piraeus Port towards Keratsini [58].

Athens Sewage Outfall (ASO).

The northern part, named the Inner Saronikos Gulf, receives the treated wastes of approximately 5 million people from the Wastewater Treatment Plant (WWTP) outfall that has been discharged south of Psyttalia Island since 1995 (Figure 3). Before that, from 1950 to 1995, the untreated domestic wastewater and part of the industrial wastewater of Athens municipality had been discharged into the shallow waters (about 30 m) of eastern Keratsini Bay, at rates of ~200,000 m³/day in the 1970s [21,23]. In 1974, a solid breakwater was also constructed, directly north of the outfall, further restricting the already limited circulation in the area [25] (Figure 3). A large sludge field covering approximately 9 and later 13 km² of the seafloor with black anoxic mud has been reported since 1975 [21,25]. Approximately twenty years later, the total sewage flow discharging to the shallow Keratsini Bay was estimated to be about 600,000 m³/day [27,59]. The effect of the sewage discharge was largely imprinted in the marine environment; it provoked a dramatic loss of clarity in the ambient water, as well as a repulsive odor and the accumulation of organic-rich sediments on the seafloor [27,59]. The industries that were discharging wastewater through the Athens Sewage Outfall during the 1970s are presented in Table S1_1 (Supplementary File_S1).

Piraeus Port.

Piraeus Port has been the most important and largest port in Greece since Archaic times, and it is among the largest in the world in terms of container and passenger traffic. Its modern facilities extend to the areas of Drapetsona, Keratsini, and Perama. It is the main unloading port in Greece and, at the same time, serves as a crucial trans-shipment point for sea, rail, and road transportation to the Mediterranean and Central Europe (Figure 1). The marine repair zone occupies the western part of the coast of Perama. Thus, due to shipping (loading and uploading, accidental spills) and urban (wastewater emissions) activities, Piraeus Port, like most harbors, has been subjected to significant environmental pollution [60] and is a pollution source for the Saronikos Gulf. The heavy metal and metalloid concentrations in its sediments are efficient indicators for the assessment of the area’s pollution.

Marine sediments derived from Piraeus Port were reported by Grimanis et al. [22], with the elevated As, Cd, and Zn values probably due to the red mud deposits from the Fertilizer Plant outside the port. Additionally, seawater and sediments were found with high levels of heavy metals, indicating that Piraeus Port may serve as a source of metals affecting the neighboring coastal zone [61]. Significant contamination by organotins in the port’s sediments and biological tissues was also recorded in 1999 [62], attributed to antifouling paints. Additionally, a correlation between the increased metal concentrations (Zn, Sn, W, Ag, Cd, As, and Fe) and the elevated levels of rare earth elements was recorded for Piraeus Port sediments by Kostakis et al. [63].

Elefsina Bay.

Elefsina Bay was named after the city lying on its coast, 20 km west of the city of Athens. It is an internal gulf in the north of the wider Saronikos Gulf. It is a semi-enclosed basin formed in an E–W direction (Figure 4). The bay area is estimated at 67 km², with a maximum depth of 33 m, enclosing approximately 1.2 km³ of water. Two narrow entrances allow water exchange with the open sea. The local tides are less than 5 cm in height, and the sedimentation rates range from 0.5 to 0.8 cm/year [64,65]. It is also documented that the lack of strong currents, anoxic conditions, and water stratification, especially during the summer, with a clearly defined thermocline, intensify the pollution problem by favoring the accumulation of sediments and inhibiting the natural destruction/decomposition of pollutants [66]. The largest industrial zone in Greece is located in the northern part of Elefsina Bay, where about 2200 companies have been estimated by the local Development Association (DATP, 2003) to be set on the Thriassion Plain (Figure 2 and Figure 4). Some of the largest industrial compounds in Greece are hosted in the region, such as two oil refineries, two steel industries, two cement factories, one industry of weapon systems, large warehouses, oil distribution facilities, three units of used lubricant processing, one paper mill, chemical industries, and industries and manufacturers of plastic products.

The heavy metal concentrations in Elefsina Bay sediments have also been studied since the 1970s [24], indicating elevated Zn values (2240 ppm) of the seafloor sediments, mainly due to the steelwork industries. The first established study on the accumulation rate was also conducted in this area by Scoullos and Oldfield [67], who recorded a value of 0.5 cm/year and characterized the upper 20 cm sediments as anthropogenically impacted. Similar conclusions were extracted by several researchers, setting the largest environmental damage at Elefsina Bay during the period 1960–1980 by the steel industries depositing untreated toxic wastes directly into the sea [68]. It has also been calculated that 20% of the sewage and outflow from the Athens Sewage Outfall reached Elefsina Bay [66]. Apart from industrial waste, environmental pressures in the bbay are also attributed to the Ag. Georgios stream (tanneries, paper mill, used lubricant processing), shipyards, scrap metal yards, and suspended atmospheric particles drained from a landfill located upstream in Ano Liossia [66].

2.2. Literature and Meta-Analysis Review Dataset and Analysis

The current review was based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). To collect data from published studies on the heavy metal pollution in marine sediments of the Saronikos Gulf, keyword searches were conducted by three researchers working independently (A. G., S. S., and L. A.) on electronic databases, including Scopus, Web of Science, Science Direct, and Google Scholar. Searches within references were also conducted (Figure 5). The literature search conducted on 5 July 2024, is described as follows: The Boolean flow was used in the initial search of the Scopus database through titles, abstracts, and keywords ((“heavy metals” OR “trace metals”) AND (“coastal sediments” OR “marine sediments”)). The Saronikos Gulf and Elefsina Bay were used to search within the results, and the time frame covered the period from 1 January 1974 to 30 June 2024 to include all possible references to these topics over the last five decades. Sixty-two documents were found. The database Web of Science was also used with the following query: (ALL = (heavy metal) OR ALL= (trace elements)) AND ALL = (Greece), resulting in 172 documents. The search was also refined searching within the results using the words “marine sediments” and “Saronikos Gulf”, returning 17 papers. A total of 120 research articles were initially identified in Science Direct using the terms “heavy metals in marine sediments from Saronikos Gulf”. Additionally, using the abovementioned terms WITH ALL THE WORDS and ANYWHERE IN THE ARTICLE, Google Scholar returned a total of 21 research articles. To minimize bias and reduce reporting errors, a stringent bibliographic screening process was applied that eliminated duplicate data sources, non-sediment studies, studies not in the English language, and studies reporting secondary data (Figure 5). Thus, 33 relevant peer-reviewed manuscripts in total were considered appropriate for inclusion in the present literature review.

The 33 selected articles are distributed across 21 journals, with most articles obtained from Marine Pollution Bulletin, Science of the Total Environment, Water, Air, & Soil Pollution, and Journal of Radioanalytical and Nuclear Chemistry (Figure 6c). In addition, the following were extracted from the articles: authors, year, specific Saronikos region, and metals under investigation. After downloading, the articles were read in full to obtain the data.

Dataset of articles used in meta-analysis review

Papers were included in the meta-analysis review if they presented quantitative and composition data on the total concentrations of heavy metal contamination in surficial marine sediments of the Saronikos Gulf and were published in English. Articles with insufficient or incomplete data on pre-established data were excluded. Additionally, papers were excluded from the study if (a) they did not analyze the quantification of heavy metal concentrations; (b) they did not provide the composition data of metals; (c) only sequential extraction methods were used; (d) the full paper was not available; and (e) they were either editorial/conference papers, reviews, or book chapters. This process resulted in the selection of 14 peer-reviewed articles. The full texts of the articles eligible for inclusion were accessed from the same databases used for the search. The data obtained were organized and processed in an Excel spreadsheet. The average metal values extracted from the 14 individual studies are used for all the analyses in this study and represent estimates of the average extent of metal pollution in the sediments (information on the reviewed research papers is presented in Table S1_2, Supplementary File_S2).

2.3. Literature Data Processing and Treatment

2.3.1. Sediment Pollution Spatial and Temporal Distribution (Meta-Analyses)

GIS analysis.

The ArcGIS Pro 3.2 software (ESRI Inc., Redlands, CA, USA) Kriging interpolation tool was used in this investigation to illustrate the levels of heavy metals present on the Saronikos Gulf seafloor. This method is commonly used to estimate the unknown spatial scale of sediment data from the known data of sites that are adjacent to the unknown sites [69].

Maps of environmental geochemistry created with GISs are frequently used to locate hotspots for sediment pollution and sediment quality. The GCS WGS 1984 coordinate system was used to create a shape file in the present study. Then, the metal concentrations, computed PLI values, and sample station locations served as the input data for the creation of symbol maps that were used to examine the distribution and bioavailability of the metals in the sediments using Surfer 11 software. Then, using ArcGIS Pro 3.2 software, the acquired maps were superimposed with additional thematic maps, such as those of land and water. Thus, in the present study, the GIS was used to (a) locate the sampling stations in the study area, (b) create geochemical maps that display the distribution of metals in the sediments per decade, and (c) visualize the overall pollution of the studied metals in different stations over the years and per decade.

2.3.2. Environmental Indices (Meta-Analyses)

Heavy metal concentrations obtained from the 14 research papers were used to estimate the ecological profile of the Saronikos Gulf per decade. The ecological risks caused by heavy metals in the sediments of the Saronikos Gulf were evaluated using the Pollution Loading Index (PLI) [70]. The PLI provides an assessment of the sediments’ overall toxicity level. Additionally, the Potential Ecological Risk Index, introduced by Hakanson [71], was used, and the human risk was calculated through ingestion and dermal contact using the Human Health Risk Assessment [20,72,73,74]. As surface sediments can act as both sinks for metals and secondary contaminants in aquatic ecosystems, it is important to understand the exposure risk of heavy metals for people (especially fishermen and consumers of local fish, from both pathways and mainly through ingestion) living in the coastal area and for the coastal reclamation [47]. Additionally, Kumar et al. [75] conducted a review of data on the heavy metals in sediments from India and performed a Human Health Risk Assessment based on these data. Wang et al. [76] evaluated the non-carcinogenic risks associated with nine heavy metals across various sediment types (mud, mud–sand, and sand) collected from seven estuarine wetlands in four coastal provinces of China. These recent studies underscore the exposure to human health risks posed by heavy metals, even in estuarine sediments, and emphasize the importance of assessing the non-carcinogenic risks linked to surface sediments in relation to human health.

The indices used in this meta-analysis study, their equations, and their classifications are summarized in

Table 1. Environmental indices used in this meta-analysis review study.

Index	Procedures of Calculation	Classification	Description	Reference
PLI	PLI = (CF1 × CF2 × CF3 × … × CFn)1/n, where CFmetals is the ratio between the content of each metal to the background values (background values according to Gkaragkouni et al. [32] in sediment; CFmetals = Cmetal/Cbackground	(0) (1) (>1)	Unpolluted sediments Baseline level of contamination Progressive deterioration of the environmental conditions and increasing pollution	[70]
PERI (RI)	PERI${{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{E}}_{\mathrm{r}}^{\mathrm{I}}\mathrm{E}}_{\mathrm{r}}^{\mathrm{i}}$ = ${\mathrm{T}}_{\mathrm{r}}^{\mathrm{i}}\times {{\mathrm{C}}_{\mathrm{f}}^{\mathrm{I}}\mathrm{C}}_{\mathrm{f}}^{\mathrm{i}}$ = ${\displaystyle \raisebox{1ex}{${\mathrm{C}}_{\mathrm{n}}^{\mathrm{i}}$}\!\left/ \!\raisebox{-1ex}{${\mathrm{B}}_{\mathrm{n}}^{\mathrm{I}}$}\right.}$ ${\mathrm{E}}_{\mathrm{r}}^{\mathrm{i}}$ is the potential ecological risk factor for a given substance, ${\mathrm{T}}_{\mathrm{r}}^{\mathrm{i}}$ is the toxic response factor, ${\mathrm{C}}_{\mathrm{f}}^{\mathrm{i}}$ is the contamination factor, ${\mathrm{C}}_{\mathrm{n}}^{\mathrm{i}}$ is the heavy metal concentration in the sediments, ${\mathrm{B}}_{\mathrm{n}}^{\mathrm{i}}$ is the background concentration of heavy metals in the sediments, and ${\mathrm{T}}_{\mathrm{r}}^{\mathrm{i}}$ is the biological toxicity factor; i.e., Cd = 40, Cr = 30, Cu = 5, Pb = 5, and Zn = 1	<150 150–300 300–600 >600	Low ecological risk Moderate ecological risk Considerable ecological risk Very high ecological risk	[71]
HHRA	$ADDing=\mathrm{C}\mathrm{S}\mathrm{E}\mathrm{D}\mathrm{I}\mathrm{M}\mathrm{E}\mathrm{N}\mathrm{T}\times \frac{\mathrm{I}\mathrm{N}\mathrm{G}\mathrm{R}\times \mathrm{E}\mathrm{F}\times \mathrm{E}\mathrm{D}}{BW\times AT}\times 10^{-6}$ $ADDder=\mathrm{C}\mathrm{S}\mathrm{E}\mathrm{D}\mathrm{I}\mathrm{M}\mathrm{E}\mathrm{N}\mathrm{T}\times \frac{\mathrm{S}\mathrm{A}\times \mathrm{A}\mathrm{F}\times \mathrm{A}\mathrm{B}\mathrm{F}\times \mathrm{E}\mathrm{F}\times \mathrm{E}\mathrm{D}}{BW\times AT}\times 10^{-6}$ HQi = ADDi/RFDi Csediment is the metal content in the sediment sample, and the RfD (mg/kg day) values used in the present study were the following: Cu: RfDing = 4.00 × 10−2, RfDder = 1.20 × 10−2. Pb: RfDing = 3.50 × 10−3, RfD der = 5.25 × 10−4. Zn: RfDing = 3.00 × 10−1, RfDder = 6.00 × 10−2. HI = ∑HQi = ∑(ADDI/RfDI)	HQ > 1 HI > 1	Risk of occurrence of harmful health effects Possible occurrence of non-carcinogenic effects	[77,78,79,80,81]

, while the analytical details on each index (including definition, exposure factors, and reference values used to estimate the intake values and health risks of heavy metals in sediments for the present study) are presented in Supplementary File_S3.

3. Results and Discussion

3.1. Literature and Systematic Review

Figure 6 summarizes the main characteristics of the thirty-three research articles studied. Notably, surficial and core sampling have been almost equally used in seafloor research over the years (Figure 6a). Additionally, very few research papers present rare earth elements, radionuclides, or only single-element values (Figure 6a). In the same graph, 7 studies that used samples from the entire Saronikos region are depicted, and 26 that focused their interest specifically on sampling near the pollution point sources. Finally, 12 of them establish background values for the Saronikos Gulf sediments. In Figure 6b, the heavy metals studied per number of research articles are presented, depicting Zn as the most studied metal and Hg as the least. Figure 6c demonstrates the scientific journals wherein the retrieved papers were published, displaying the Marine Pollution Bulletin as well as the Journal of Radioanalytical and Nuclear Chemistry as the most preferable. Finally, Figure 6d displays the number of publications (%) per decade, revealing the 1980s as the most productive decade for the topic of metal pollution in the Saronikos Gulf, followed by the 2000s and 2010s. It can be also concluded that the researchers’ interest between 1974 and 2024 was approximately steady, without great fluctuations. More analytically, the retrieved research papers categorized per decade of publication revealed the following.

1970s.

The research on heavy metals in Greece begins with the study of Saronikos Gulf marine sediments in 1972, when 84 surficial sediment samples were collected and analyzed employing sequential extraction and the use of INAA at the Democritus Institute. The research though was mostly targeted around the sewage outfall and the fertilizer company, focused on defining the extent of the sludge field formed [22,23,25,82]. Sediment cores were also collected at Elefsina Bay and analyzed for their metal concentrations and magnetic parameters [24]. The studies conducted in the 1970s set the ground for research in this field in Greece and provided useful conclusions, but they lacked data availability, since, in all cases, only maximum–minimum or average metal values were published. Griggs and Hopkins [25] also presented the elevated values of organic carbon around the outfall and used it to estimate the sludge field cover area from about 9 to 13 km² and the overall area affected by pollution to an extent of 160 km². Another interesting finding is also that the heaviest pollution source at that time was attributed to the Fertilizer Plant, since its sludge was richer in metals than the Athens Sewage Outfall (probably due to the early stages of the urbanization and industrialization of the city of Athens), recording As values of 1500 ppm and Zn values of 2992 ppm. Piraeus Port also appears to have been heavily polluted by most metals, while the ASO was much enriched in Cr, and Elefsina Bay had high levels of heavy metals, but to a lesser extent than the other two point sources. The most studied metals in the decade were Zn, Hg, Sb, Cr, and As; however, the C_org, Ag, Fe, Mn, and Hg values were also studied. It could also be noted that Hg (with maximum values of 10.1 ppm, sampled near the FP) appears again in the research in the 2000s by [83], recording lower values than the FP area but higher values than those measured near the ASO and Elefsina Bay (0.5 ppm).

1980s.

In the 1980s, the area of the Saronikos Gulf stayed in the spotlight of scientific heavy metal research, with an emphasis on its northern part. Many researchers used surficial and core sediment samples to study heavy metal concentrations [41,42,46,58,64,84] and rare earth element (REE) concentrations [84]. Apart from Voutsinou-Taliadouri, 1981 [42], which presents 87 samples from the broader area of the gulf with a focus on the most polluted spots (the Athens Outfall, Piraeus Port, and the Fertilizer Plant), two of the papers focus on Elefsina Bay [41,64], and four of them focus on the central sewer outfall area [46,58,84,85]. The Fertilizer Plant seems to have contributed the most to the pollution of the area in Zn and Fe among the three point sources of the area: the Fertilizer Plant, the Athens Sewage Outfall, and Piraeus Port (FP, ASO, and PP). At Elefsina Bay, there was another hotspot for pollution: the iron and steel works, but, according to Scoullos [41], there is also a spot with a high concentration of heavy metals in the bay influenced by the ASO (near the eastern channel). Although there was an increased interest in studying the persistent, incremental pollution in the Saronikos Gulf in the 1980s, some studies present only single metals (e.g., Zn or V), or they just present the average, minimum, and maximum values of the studied elements, or even when it comes to data obtained from sediment cores, researchers only present downcore variation diagrams. The published studies [58,85] include the data matrixes of the metal values calculated in the papers at the end of this decade. It is worth noting that, in this decade, rare earth elements (La, Ce, Sm, Eu, Tb, Dy, Yb, Lu) were estimated [84], and the Pb concentrations in sediments from Elefsina Bay were calculated, with Scoullos [64] recording elevated values (max Pb = 600 ppm).

1990s.

During the 1990s, research interest in Saronikos, apart from the areas near the pollution point sources, included the marinas (Alimos, Vouliagmeni, Zea) [62,86] and studies on elements (apart from heavy metals) such as rare earth elements and organotins [62,87]. Apart from Grimanis et al. [88], who studied 35 sediment samples, there was not a dense sampling network involved in the Saronikos Gulf. According to Kokovides et al. [86], Zea is the most polluted by Fe, Cr, Zn, Cu, Ni, Mn, Pb, and Cd among the three studied marinas due to anthropogenic activities, while metal removal from the dredged material was also studied [89]. Metal values were available as minimum, maximum, and mean values, while Ag and Sb were also studied [88], presenting higher concentrations near the Fertilizer Plant than those near the Athens Sewage Outfall.

2000s.

Total heavy metal concentrations using sediment cores and surficial samples from Piraeus Port were obtained by Sakellariadou et al. [61]. Other sampling subareas in the gulf were Elefsina Bay, Keratsini Bay, and the marinas in East Saronikos [43,66,83,90,91]. Surficial (the vast majority) and core samples were obtained [65,91]. Additionally, in this decade, the coast of Ag. Kosmas was also studied by Karageorgis et al. [90], determining the heavy metals and Sr values in the sediments, whilst the full table of the metal values of this study was also published. Elefsina Bay was still in the spotlight of the research in the 2000s [65,66], for which both surficial and core samples and sediments were collected, reporting values of heavy metals in the bay’s sediments larger than those of most Aegean ports and industrial coasts in Greece. Mavrakis et al. [66] also concluded that high Ni, Cu, Fe, and Mn values were directly associated with industrial activity (e.g., Skaramaga shipyards), but they noticed a reduction in the Mn and Cu concentrations in the most recently deposited sediments. Another important aspect in this decade was also the study of several marinas of Saronikos (Zea, Mikrolimano, Alimos, Glyfada, Voula, Vouliagmeni, and Lavrio) in terms of their heavy metal concentrations [91], determining the total metal values. Additionally, Tapinos et al. [91] underlined the need for environmental protection policies in the marinas. Moreover, Galanopoulou et al. [43] reported elevated heavy metal values in Keratsini harbor, especially for Cd and Zn. It should be noted that, in this decade, the researchers’ datasets were being published and were thus available and accessible.

2010s.

The main characteristic of the publications in the 2010s is their focus on specific regions (pollution hotspots) of the Saronikos Gulf, Elefsina Bay, and the Psyttalia–Keratsini straits [45,92,93]. Elevated concentrations of many elements were calculated (Al, As, Ca, Cu, Cr, Fe, Mo, Mg, Mn, Ni, P, Pb, Sb, Sn, Si, S, Ti, V, and Zn [45]), as well as radionuclides [92]. Then organic contamination of the surficial sediments derived from Keratsini was also studied in this decade [44]. The obtained data that were retrieved in this decade indicated the necessity of the continuous monitoring of the heavy metal pollution in Saronikos Gulf sediments.

2020s.

Karageorgis et al. [26] analyzed 216 surface sediments samples from 68 sites, collected from 1999 to 2018, for their major and trace elements and basic physical parameters. Results revealed that Elefsina Bay, the Inner Saronikos Gulf, and some areas of the Outer Saronikos Gulf are contaminated by Cu, Zn, and Pb. Apart from the study by Karageorgis et al. [26], which constituted the most thoroughgoing and systematic research in the area, the retrieved papers were centered on specific regions of the gulf, Elefsina Bay, and the Athens Sewage Outfall [27,30,31,32]. Prifti et al. [30] studied the temporal evolution and chemical speciation of eleven elements in sediment cores from Elefsina Bay and the Inner Saronikos Gulf, emphasizing V and Ag. The results revealed the extensive pollution of the sediments by Ni, Cr, Cu, Zn, As, Mo, Cd, and Pb from the 1910s and 1960s in eastern and western Elefsina Bay, and a significant decrease in the sediment enrichment in V, Ni, Cr, Cu, Zn, As, Cd, Pb, and Ag since 2000 in the part of the Inner Saronikos Gulf that is mainly influenced by the WWTP of Athens. Finally, in the Inner Saronikos Gulf, the Ag sediment enrichment was found to be remarkably high. Moreover, Dimiza et al. [28] studied the living benthic foraminifera composition and its relation to environmental parameters such as the grain sizes, organic carbon contents, and heavy metal concentrations from the surficial sediment layer collected in the Elefsina Bay and the Inner Saronikos Gulf, estimating the heavy metal (Cu, Cr, Ni, Pb, Zn, and As) contents. Sediment cores derived from the organic mud between Psyttalia and Keratsini analyzed for their heavy metal contents revealed extremely elevated concentrations of Cu, Zn, Pb, and C_org in the sludge [27]. Additionally, Gkaragkouni et al. [32] studied sediment cores in terms of their lithologies, organic carbon contents, and heavy metal concentrations in the area between Perama and the island of Salamina. The research concluded that the upper 40 cm layer of the sediments in the strait are enriched in the metals Ag, As, Cd, Cu, Ni, Pb, Sn, and Zn, of anthropogenic origin, and have been introduced into the marine environment over the last 65 years. This decade is characterized by a large number of publications with available datasets, emphasizing ecological assessment and sediment chronology.

Summarizing the abovementioned, the earlier published studies conducted in the Saronikos Gulf (1970s–1980s) set the ground for the research in this field in Greece and provided useful conclusions, but they lacked data availability, since, in all cases, only the maximum–minimum or average metal values were published. Additionally, the main pollution source in the area was the FP, whilst the organic carbon around the ASO that was used to estimate the sludge field covered an area from about 9 to 13 km². Moreover, Elefsina Bay was highlighted in the 1980s as another pollution hotspot in the area. In the 1990s, the marinas of Alimos, Vouliagmeni and Zea were also being studied, while the FP and ASO were still reported as the main pollution contributors in the area. In the most recent decades (2000–2010–2020), the research has expanded in the gulf and more data are available, whilst many more elements have been estimated. Very high heavy metal concentrations were also estimated in Keratsini harbor (2000s). The most thoroughgoing and systematic research in the area was reported by Karageorgis et al. [26] in the 2020s, whilst the trends from several studies converge, reporting slightly reduced heavy metal concentrations in the most recently deposited sediments, also highlighting the need for further systematic monitoring in the gulf employing more elements.

Instruments used in reviewed metal analysis studies in Saronikos Gulf.

The instruments most used in the metal analyses during the surveys of sediments of the Saronikos Gulf, with their characteristics and analytical performances, are presented in

Table 2. Instruments used in metal analyses of sediments of Saronikos Gulf.

	Liquid Sample	Solid Sample	Sample Vol. (mL)	Max. Matrix Conc. (g/L)	Detection (ng/mL)	Detection (ppm)	Sequential Multielement	Simultaneous Multielement	Matrix Effects	Spectral Interferences	Precision % RSD
INAA	Possible	Possible	0.01	70	0.01–100	0.05–50	Yes	Yes	Small	Few	5
ICP-AES	Ideal	Possible	1–10	10–100	0.1–10	-	Yes	Yes	Small	Large	0.5–1
Flame-AAS	Ideal	(a)	5–10	30	1–103	-	Possible	No	Large	Few	0.5–1
GF-AAS	Ideal	Possible	0.01–0.1	200	10−2–0.1	-	Possible	Yes	Moderate	Few	3–10
ICP-MS	Ideal	Possible	1–10	0.1–0.5	10−3–10−2	-	Yes	Yes	Moderate	Significant	1–3
WD-XRF	Possible	Ideal	(b)	(b)	(b)	0.1–104	Yes	Yes	Large		1

Note(s): INAA: Instrumental Neutron Activation Analysis; Spark-AES: Spark Emission Spectrometry; Flame-AES: Flame Atomic Emission Spectrometry; ICP-AES: Inductively Coupled Plasma Atomic Emission Spectrometry; Flame-AAS: Flame Atomic Absorption Spectrometry; GF-AAS: Graphite Furnace Atomic Absorption Spectrometry; ICP-MS: Inductively Coupled Plasma Mass Spectrometry; WD-XRF: Wavelength Dispersive X-ray Fluorescence Spectrometry; (a) not applicable, (b) depends on the analytical problems or the sample preparation.

.

According to Table 2, the INAA method was the most widely used analytical technique in the early 1970s and 1980s in the Saronikos Gulf research. However, in the 1990s and early 2000s, AAS (flame or flameless) became the most popular method, and in the most recent studies, the ICP-MS, ICP-AES, and XRF systems were employed. On of the advantages of INAA, apart from being a relatively cheap analytical method, is that it is nondestructive; hence, the same sample can be used for other measurements, the sample size can be as little as one milligram, the detection limits for many elements are in the nanogram range, no chemical preparation is required, and finally, 40 elements can be measured essentially simultaneously. The major disadvantage of INAA is that not all elements of interest can be analyzed by it. X-ray fluorescence (XRF) analysis, which, in most cases, is also nondestructive and suitable for solids, liquids, and powders, has been used in the most recent analyses, but the combination of the methods (INAA and XRF) can result in high-quality data for about 60 elements in the periodic table [94].

Atomic Absorption Spectroscopy (AAS) has also been used for heavy metal determination in sediments from the Saronikos Gulf, mostly in the 1990s and 2000s. The AAS method is now being replaced mostly by ICP-MS due to the disadvantages of AAS, as it is mostly used for liquids rather than solids since the substance must be vaporized before analysis. Additionally, the techniques, which do allow for solid-substance testing, cannot be used on non-metals. In addition, the results may be affected and interfered with by additional substances present in the sample or the surrounding air. Thus, another one of the most recently used methods is Inductively Coupled Plasma Mass Spectrometry (ICP-MS). Even though, in this method, many sample preparation steps are involved, and it includes high costs, ICP-MS’s major advantages are that it can be used to determine elemental concentrations for most of the elements in the periodic table and it is highly sensitive. The evolution of the analytical systems led to the estimation of larger numbers of heavy metal concentrations in the most recent studies. Thus, more elements contributing to pollution can be determined and taken into consideration when planning environmental policies. However, the differentiation in the analytical methods can cause difficulties in comparisons between research results. A statistical comparison could prove useful in this direction. A comparison between AAS and ICP-MS [27] showed a high level of correlation (R > 0.7) for both methods using the exported concentrations of Cu, Pb, Ni, and Mn. It is also worth mentioning that a comparison between ICP-MS and γ-ray spectrometry was conducted by Papaefthymiou et al. [92] using uranium concentrations in the same samples, proving a relatively strong linear correlation (R = 0.893) between the two methods.

At this point, it should also be noted that among the metal analyses, the total digestion was mostly used among the studies, although sequential extraction was also used [42,65], or it was combined with the total digestion in numerous studies over the years [58,85,88]. While the total digestion process considers not only the human-induced concentrations of elements but also their background levels, the outcomes of partial or sequential extraction methods are defined operationally, making them less straightforward for comparison. Furthermore, the different fractions extracted from different sequential methods are not univocally related to the anthropogenic fraction [95]. The total digestion was used in the vast majority of the studies in the Saronikos Gulf, in the fine sediment fraction (from <55 μm in the earlier studies to <62 μm in the most recent ones).

Background concentrations estimated in preview studies.

According to Birch [1], establishing background metal concentrations in marine sediments is essential to estimate the extent of the contamination, anthropogenic contributions, and progress due to management objectives. Filzmoser et al. [96] defined background concentrations as the properties, locations, and distributions of geochemical samples that represent the natural variation in the material being studied in a specific area that is not influenced by secondary processes, such as chemical forming (diagenetic) processes or anthropogenic contamination. Setting metal background concentrations is indispensable since marine sediments serve as environmental indicators, and environmental indices are more often used. Background metal values were used in the Saronikos Gulf for various elements in ten of the previous studies. The methods employed were empirical, such as the use of the global mean concentrations, pristine areas within the same ecosystem, and sedimentary cores. According to Birch [1], the use of sediment cores is particularly appealing since it does not only provide background levels but also creates a record of contamination and, if dated, establishes the relaxation rate and the age at which the pollution began.

Griggs et al. [23] used background values derived from unpolluted areas in the Saronikos Gulf collected further away from the pollution point sources. The established values fluctuate for C_org, As, Co, Cr, Sb, Zn, and Hg as follows: ~1%, 5–7 ppm, 4.5–9 ppm, 50–70 ppm, 0.3–0.7 ppm, 40–80 ppm, and 0.33–0.46 ppm, respectively (Table S2_1 Supplementary File_S1). Additionally, Scoullos et al. [41] studied the Zn concentrations in Elefsina Bay and reported background concentrations between 60 and 80 ppm based on bottom sediment cores of a 1 m length. In the study by Voutsinou-Taliadouri [42], which was based on surficial samples collected away from pollution sources, background values were estimated for C_org (0.2%), Cu (1 ppm), Fe (2 ppm), and Zn (1 ppm). Angelidis and Grimanis [58] calculated the background values of the elements C_org, Co, Cr, Fe, Sb, and Zn. Their calculations were based on the mean values of three samples collected about 8 km from the ASO and were estimated as follows: (1–2)% C_org, 8–9 ppm; Co, 160–220 ppm; Cr, 14–17 mg/g; Fe, 6 ppm; V and Zn, 100–120 ppm. Kalogeropoulos et al. [46] used a mean of 30 ppm as a background value for vanadium (V) using sediment samples collected further away from the known pollution sources. In a similar direction, Kokovides et al. [86] calculated the natural metal contents using the average metal values extracted from samples collected “in theoretically unpolluted areas” of the gulf (Table S2_1 Supplementary File_S1). Kalogeropoulos et al. [87] also published background values using samples collected northeast from the pollution sources for the following elements: Ag, As, Co, Cr, Fe, Mn, Sb, Sc, V, and Zn (Table S2_1 Supplementary File_S1), while, a year later, Grimanis et al. [88] reported mean concentrations of a “clean” core of Ag and Sb at 0.2 ppm and 0.59 ppm, respectively. Karageorgis et al. [26] used dated cores to determine preindustrial sediments and establish background concentrations collected from the western, outer, and inner Saronikos and Elefsina Bay, determining the differentiation among these subregions; the values are presented in Table S2_1 Supplementary File_S1. Prifti et al. [30] calculated the background values, obtaining concentrations of the metals from the preindustrial sediment layers, as defined by the calculated sedimentation rates presented in Table S2_1 Supplementary File_S1. Additionally, Gkaragkouni et al. [32] calculated the background values using the mean concentrations of bottom sediment samples collected from gravity cores collected between Perama and Salamina (Table S2_1 Supplementary File_S1).

According to

Table 3. Minimum, maximum, and average background values as estimated by various researchers in the Saronikos Gulf (metal values in ppm and Fe, C_org, and Al in %).

Location	Corg	Ag	Al	As	Cd	Co	Cr	Cu	Fe	Mn	Ni	Pb	Sb	Sc	V	Zn	Hg
Min.	0.2	0.04	1.46	5.0	0.1	4.5	8.5	1.0	0.71	95	35.7	5	0.3	5.9	16.2	1.0	0.33
Max.	0.49	0.7	7.45	25	3.5	12.3	220	25	3.5	520	115	48	0.9	6.7	90	120	0.46
Mean	0.92	0.40	3.62	12.67	0.95	8.48	106.44	11.78	4.68	258.4	63.9	23.0	0.59	6.2	43.7	62.9	0.40

and Table S2_1 Supplementary File_S1, background values have been established for Zn Cr, Co, Fe Cu, Ag, As, Mn, Pb, Ni, V, C_org, Al, Cd, Sb, Sc, and Hg. As, Zinc, Cr, Co, Fe, and Cu are among the most studied elements, and the area background values have been established for them in many studies. However, there are fewer records of background values for the less studied elements in the gulf (Ag, As, Mn, Pb, Ni, V, C_org, Al, Cd, Sb, Sc, and Hg). The minimum, maximum, and average background values for the studied metals in the Saronikos Gulf are also presented in Table 3. According to this, and due to the complexities of determining a single background value for each element for the whole area, it might be useful to obtain normalized background values from core bottom samples rather than surficial samples from areas of the gulf unaffected by pollution.

Dating techniques used in previous studies and established sedimentation rates.

Scoullos and Oldfield [67] made a first attempt to establish the sedimentation rates in Elefsina Bay and suggested that by using frequency-dependent susceptibility measurements, it is possible to establish both the anthropogenic origin of the recently deposited sediments and the source of the magnetic mineral assemblages in the suspended particles and the sea bottom sediments (i.e., the industrial emissions from the northeast section of the gulf). By combining the trace metal results with the magnetic measurements, a contemporary sedimentation rate of about 0.5 cm/year was estimated for Elefsina Bay. Additionally, Kanellopoulou et al. [83] estimated the sedimentation rate in Elefsina Bay using the ²¹⁰Pb chronology method and CRS model at about 0.29 cm/year. Another study for Elefsina Bay was conducted by Panagiotoulias et al. [45], which estimated the maximum sedimentation rates by using the ²¹⁰Pb method at 0.4 cm/year for one and at 1 cm/year for a second core. The differentiation in the sedimentation rates of the two sites was attributed to the increased sediment load inputs to the one site derived from the adjacent stream and to an artificial channel connected to the iron and steel industry. Karageorgis et al. [26] used the ¹⁴C and ²¹⁰Pb chronology methods and established an average sediment accumulation rate of 0.26 cm/year in Elefsina Bay, and an average value of 0.9 cm/year was established for Keratsini Bay in the Inner Saronikos Gulf. On the contrary, in the western basin, the sediment accumulation rates were estimated much lower, around 0.006 cm/year. Prifti et al. [30] also established sedimentation rates in Elefsina Bay at 0.11 and 0.25 cm/year and in the Inner Saronikos Gulf at 0.62 and 0.08 cm/year also using the ²¹⁰Pb chronology method and applying the Constant Rate of Supply (CRS) model.

Additionally, for the area between Perama and Salamina, the age estimations of Gkaragkouni et al. [32] were based on the extrapolated sedimentation rates derived from the radiometric activities. In the study, both the constant rate of the ²¹⁰Pb supply (CRS) and the regression/constant initial concentration (CIC) models were used for quantifying the sedimentation rates from the decreasing trend of unsupported ²¹⁰Pb with the depth, estimating the mean sedimentation rate for the inner Saronikos Gulf at 0.45 cm/year.

Table 4. Sedimentation rates estimated in Saronikos Gulf.

Subarea of Saronikos Gulf	Sedimentation Rate (cm/year)	Reference
Perama–Salamina strait	0.45	[32]
Inner Saronikos	(0.62 and 0.08)	[30]
Elefsina Bay	(0.11 and 0.25)	[30]
Wester basin of Saronikos	0.006	[26]
Inner Saronikos	0.9	[26]
Elefsina Bay	0.26	[26]
Elefsina Bay	(0.4–1.0)	[45]
Elefsina Bay	0.29	[83]
Elefsina Bay	0.5	[67]
Min.	0.006	Saronikos
Max.	1.0	Saronikos
Mean	0.41	Saronikos

presents the sedimentation rate variations among the Saronikos subregions, with the lower rates noted in the western basin and the greater recorded in the Inner Saronikos and Elefsina Bay. The average value of the sedimentation rate in Saronikos Gulf according to previous studies has been calculated at 0.41 cm/year. Based on these values, the thickness of the contaminated sediments is expected to be about 25–30 cm in the Inner Saronikos Gulf, Elefsina Bay, and the Perama–Salamina strait, taking into consideration that urbanization and industrial growth in the area occurred in the past 50 years. Establishing background (preindustrial) values is of great importance, since they can be used in efficient ecological assessments.

Contamination assessments (environmental indices) used in previous studies

According to Birch [1], among the indices used in the assessment of the sediment conditions, the Enrichment Factor (EF) has been the most widely used and has been a successful enrichment indicator since 1974. This index became commonly used by the 1990s, employing local and global values as normalizing agents for Al, Fe, and other elements [1]. The Geoaccumulation Index (I_geo) is another commonly used index employing background values, preferably derived from local or regional areas [1]. Similar trends are observed in Saronikos Gulf environmental studies, since the EF and I_geo are the most used indices, followed by the PLI (

Table 5. (a) Environmental indices employed in the evaluation of sediment contamination in the Saronikos Gulf. (b) Indices used in previous studies to estimate the potential biological effects of heavy metals in the Saronikos Gulf.

Environmental Index	Equation/References	Metals Employed	Studies Referenced
(a)
Micropollutant Index	$log{\displaystyle \frac{\mathrm{A}s\times \mathrm{C}\mathrm{r}\times \mathrm{H}\mathrm{g}\times \mathrm{S}\mathrm{b}\times \mathrm{Z}\mathrm{n}}{\mathrm{A}s\mathrm{o}\times \mathrm{C}\mathrm{r}\mathrm{o}\times \mathrm{H}\mathrm{g}\mathrm{o}\times \mathrm{S}\mathrm{b}\mathrm{o}\times \mathrm{Z}\mathrm{n}\mathrm{o}}}$	As, Cr, Hg, Sb, Zn maximum pollution in Keratsini Bay near the outfall and outside Piraeus harbor	[21,22]
Geoaccumulation Index	[38]	Heavily polluted by Cd, Pb, W, As, Se, and Zn and highly contaminated concerning Cu and Cr	[43]
Contamination Factor (Cf) Modified Contamination Degree (mCd)	CF = (metal content in polluted sediment/background value of the metal) [71] $\mathrm{mCd} = {\displaystyle \frac{{\sum }_{i=1}^{n}CFi}{n}}$ [39]	Al, As, Ca, Cu, Cr, Fe, Mo, Mg, Mn, Ni, P, Pb, Sb, Sn, Si, S, Ti, V, and Zn: high	[45]
Enrichment Factor (EF)	EF = (X/C sample)/(X/C reference) (IAEA 1992)	As, Br, Se, Sb, and Zn are the most enriched elements in most sites	[92]
Enrichment Factor (EF) Modified Pollution Index (MPI)	$\mathrm{EF} ={\displaystyle \frac{\frac{Element}{Al}sample}{\frac{Element}{Al}background}}$ [37] $MPI=\sqrt{{\displaystyle \frac{{\left(EFmean\right)}^2+{\left(EFmax\right)}^2}{2}}}$ [98]	Cu, Zn, Pb > 10: “very severe modification” (Inner Saronikos) V, Cr, Mn, Co, Ni, Cu, Zn, As, and Pb: heavy to moderately heavily polluted (Elefsina Bay)	[26]
Enrichment Factor (EF)	[37]	Ag, V, Cr, Mn, Fe, Cu, Zn, As, Mo, Cd, and Pb: “severe modification” (Inner Saronikos)	[30]
Geoaccumulation Index (Igeo) Enrichment Factor (EF) Pollution Loading Index (PLI)	[38,70,99]	Cu > Zn > Mo > Ag > Cd > Cr > Pb: “extremely to moderately polluted” Cu > Zn > Mo > Ag > Cd > Cr > Pb: “high to medium enrichment” > 1 contaminated sediment	[27]
Enrichment Factor (EF) Geoaccumulation Index (Igeo) Pollution Loading Index (PLI)	[37,38,70]	Ag < Pb < Sn < Zn < Cu: very high to significant enrichment Ag, Pb, Sn, Zn, Cu: moderate to heavy contamination > 1 contaminated sediment	[32]
(b)
Corresponding sediment quality guidelines (SQGs), effect range low/effect range median (ERL/ERM)	[100]	Cd, Pb, As, Zn, Cu, and Cr in most of the sediments exceed the toxic effect range	[43]
SQG Mean Effect Range Median quotient (m-ERM-q) values	[100]	“Medium-to-high risks” of toxicity	[45]
Threshold Effect Level (TEL) Probable Effect Level (PEL) Risk Assessment Code (RAC)	[101,102]	Ag, Cr, Cu, Zn, As, Cd, and Pb: medium to high toxicity risk (Inner Saronikos) Cd: high toxicity risk (Elefsina)	[30]
Potential Ecological Risk Index (RI) Mean Effect Range Median quotient (m-ERM-q) values	[71,100]	As, Cd, Cr, Cu, Pb, and Zn: “low to considerable ecological risk” Zn, Cu, Pb, Ni, Cr, Cd, As, and Ag “medium to high-medium priority sites”	[32]

a). The environmental indices used by researchers in the Saronikos Gulf and the metals that were applied are summarized in Table 5a,b. As can be seen, sediment contamination indices were used from the 1970s, but in more recent studies, more environmental indices are employed to conduct ecological risk assessments in addition to evaluating the sediment condition. The Micropollutant Index was the first index used as an attempt to combine or average out the distribution patterns of individual toxic trace elements. Galanopoulou et al. [43] studied the contamination of sediments based on the I_geo. Panagiotoulias et al., the authors of [45], studied sediment samples collected from Elefsina Bay using the Contamination Factor and the Modified Contamination Degree (mCd). Karageorgis et al. [26] used the Enrichment Factor and the pollution indices of the Modified Pollution Index (MPI) (Table 5a), (i) recording the maximum values of the EF for Cu and Zn at Elefsina Bay, followed by the Keratsini area, and (ii) concluding that Pb exhibits very high EFs all over the Saronikos Gulf, with the maxima in the Inner Saronikos, and the MPI following similar trends. It should be noted that environmental indices have more often been used in most recent studies than in the earlier ones, whilst more indices are being introduced as environmental geosciences keep growing. According to Birch et al. [97], mCd is one of the indices that provides the best measurement of enrichment, whilst the Enrichment Factor is recommended for total sediment chemical data. Additionally, although the Igeo and PLI indices have been extensively used, there is the chance of underestimating the enrichment [97].

Potential Biological Effects of heavy metals used in previous studies

To determine potential adverse biological effects, Galanopoulou et al. used the corresponding sediment quality guidelines (SQGs), while Panagiotoulias et al. [45] studied sediment samples collected from Elefsina Bay using the Mean Effect Range Median quotient (m-ERM-q) (Table 5b). Panagiotoulias et al. [45] concluded that in the most recently deposited sediments (2003 to 2013), the degree of contamination is reduced compared to that in previous decades (1963–2002). Additionally, Prifti et al. [30] emphasized the SQGs, and two concentrations were defined: (i) the Threshold Effect Level (TEL) and (ii) the Probable Effects Level (PEL) concentrations (Table 5b). The Risk Assessment Code (RAC) was also used in the study to assess the mobility of heavy metals and their potential health risks based on the total concentration of the metals and their chemical speciation (Table 5b).

Moreover, sediment quality guidelines (SQGs) are also used (in the absence of information on the bioaccumulation, toxicity, and change in the structure of biological communities) to assess the risk of adverse effects posed by anthropogenic sedimentary chemicals on benthic populations [103,104,105].

Statistical analysis used in previous studies.

Statistical analysis plays a crucial role in environmental sciences, enabling researchers to comprehend environmental challenges by investigating and formulating potential solutions to the issues they examine. The utilization of statistical techniques in this field is extensive and diverse. In this direction, statistical analyses have been employed in heavy metal studies in the Saronikos Gulf mostly since the 1990s, aiming to investigate metal pollution sources and their results (

Table 6. Pollution source apportionment derived from statistical analyses employed in reviewed studies.

Group of Heavy Metals	Total Variance (%)	Pollution Source	Saronikos Subarea	Reference
As-Sb-Ag-Au-Zn-REE-	84.4%	Fertilizer Plant, ASO	Inner Saronikos	[87]
Zn-Pb-Cr-W-As-Corg	-	ASO, ports	Keratsini	[43]
Fe-Mn-Zn-Pb-Cu-Sb-S-Mo-As-Mg-TOC	-	Steel mill	Elefsina	[45]
Sn-V	-	Crude oil refineries	Elefsina	[45]
U-Corg-Silt-Clay	19.98%	FP–ASO–agriculture	Psyttalia–Keratsini	[92]
Cu-Zn-As-Pb-Corg	30.6%	WWTP, ports, industries	Inner Saronikos	[26]
Ag-Cd-Cr-Cu–Zn-Pb-Mn-Corg Fe-As-Pb–Cd-Mn Co–Ni-Mo-Mn-Cu	57.73% 15.41% 7.59%	ASO (sludge)	Keratsini–Psyttalia	[27]
Ag, As, Cd, Cu, Pb, Zn, Sn, Ba-Corg	43.26%	Industry, shipyard, ASO, WWTP	Perama–Salamina	[32]

Note(s): ASO: Attica Sewage Outfall; WWTP: Wastewater Treatment Plant.

). The descriptive statistics that demonstrate the association and correlation between variables are being reviewed. Kalogeropoulos et al. [87] used hierarchical cluster analysis, aiming to identify the partitioning processes. By this approach, Kalogeropoulos et al. [87] concluded, among others, that there was a distinction between different pollution sources and their interaction (Table 6). Another study employing statistical treatment (Pearson correlation) in their dataset was published by Galanopoulou et al. [43], aiming to demonstrate the correlation between C_org and the studied heavy metals, concluding that C_org, Zn, Pb, Cr, W, and As were significantly correlated with each other, indicating a common origin, probably the Athens Sewage Outfall (Table 6). Additionally, Panagiotoulias et al. [45] applied multivariate statistical analysis to sediment samples derived from Elefsina Bay, using R-mode and Q-mode hierarchical clustering analysis. The aim was to investigate the distribution and sources of the pollutants, as well as to discover similarities and/or dissimilarities among the parameters and samples. The use of statistics pointed out the metals associated with the steel-making process and those that are related to terrestrial inputs and other pollution sources in the area (Table 6). Panagiotoulias et al. [45] also used statistics (Q-mode HCA analysis) combined with the results of ¹³⁷Cs chronology to estimate the potential correlation among the periods indicated. A statistical approach was also conducted by Papaefthymiou et al. [92] for element and radionuclide values calculated in sediment samples collected between Psyttalia and Keratsini. In the study, statistical comparisons involved the use of One-Way Analysis of Variance (ANOVA) and correlation studies (Spearman’s two-tailed) between the measured parameters. The R-mode factor analysis was performed to investigate the inter-relationships among the C_org, U, and grain size characteristics. The results indicated that U is not only related to the sand but also to the clay fraction, suggesting two different sources and processes (Table 6). Additionally, Karageorgis et al. [26] used Principal Factor Analysis with Varimax rotation to group clay, organic carbon, and geochemical elements with common geochemical behaviors, leading to a three-factor model. The results indicated a group of C_org, Cu, Zn, As, and Pb, representative of the anthropogenic factor (Table 6) that introduces organic loads and trace element contaminants into the area. Multivariate statistical analysis (R-mode and Q-mode factor analysis) was also performed by Gkaragkouni et al. [27] on a dataset of metals and C_org for samples collected in the Psyttalia–Keratsini strait to investigate the interrelationships between metals, also exploring their origins, geochemical behaviors, and/or possible common sources. Combined results lead to the temporal variation in the prevailing pollution processes during the deposition of organic mud. Gkaragkouni et al. [32] employed multivariate statistics (R-mode factor analysis) integrated with the chronological framework for the surficial sediments of the studied cores (upper 40 cm) to reveal the pollution history of the broader area of the Perama–Salamina strait and identify several different sources of the heavy metals reaching the seafloor. The results led to the discrimination of three factors that describe the interrelations among the variables (heavy metals, C_org, Al, Ba, and Ca). One metal group (Ag, As, Cd, Cu, Pb, Zn, Sn, Ba) was attributed to anthropogenic origin entering the marine environment over the last 65 years, when industry, shipyard, and city growth took place.

Conclusively, it could be noted that sophisticated statistical analysis in the studies of Saronikos Gulf metal pollution has been employed for over 35 years to interpret the interrelationships among the metals, organic carbon content, sediment fractions, and other elements. The use of statistical analysis tends to increase in the most recent studies as data become more complex, abundant, or cover a larger area. Additionally, it can be mentioned that according to Table 6, in the 1970s and 1980s, the main pollution source in the area was the FP, introducing mainly red mud industrial wastes rich in As-Sb-Ag-Au-Zn and REE, whilst since the mid-1980s and until the mid-1990s, the main pollution source was the ASO, introducing mainly municipal waste rich in C_org, Cu, Zn, Ag, Cd, As, and Pb and industrial waste to a lesser degree rich in Fe, As, Pb, Cd, Mn, Co, Ni, Mo, Mn, and Cu. Furthermore, in Elefsina Bay, TOC-Fe-Mn-Zn-Pb-Cu-Sb-S-Mo-As-Mg entered the coastal environment due to steel mill operations, as well as Sn-V, probably due to crude oil refineries. Possible anthropogenic U inputs in the marine environment have been attributed to the FP and/or ASO. Finally, port and shipyard activities have also been identified as pollution sources mainly since the late 1980s, introducing As, Cd, Cr, Pb, Zn, Sn, W, Ni, and V into the Saronikos Gulf (Table 6).

Thus, even from the early decades of the research, the scientific community realized that marine pollution from heavy metals, apart from being a local problem, could and would affect the Saronikos Gulf on a larger scale. The effect is larger due to multi-point and non-point pollution sources, making the spread of research and sampling in the region of the gulf crucial. Therefore, a thick network of sampling stations during constant timely sampling and the use of similar analysis techniques could help with comparisons and a better understanding of the pollution evolution, sources, trends, and mechanisms in the area.

3.2. Meta-Analysis Review

3.2.1. Heavy Metal Distribution in Saronikos Gulf per Decade

The spatial distribution maps of the representative metals per decade (1970s, 1980s, 1990s, 2000s, and 2010s) are presented in Figure 7, Figure 8, Figure 9, Figure 10, and Figure 11, respectively, according to the available data retrieved from 14 research articles. The maps were constructed based on the year of sampling; thus, the last represented decade is the 2010s. In Figure 7, the distributions of C_org and Zn are presented in Elefsina Bay and Inner Saronikos in the 1970s. The maximum concentrations of C_org are detected near the ASO. The maximum Zn concentrations are also detected closer to the coast of Elefsina due to inputs from the local point sources. The dispersion of C_org through the channel of Perama–Salamina could also be remarked upon, directed from the ASO to Elefsina Bay due to the water circulation in the gulf and/or other secondary sources from the surrounding coastal region.

In Figure 8, the surficial distribution of (a) As, (b) Cr, and (c) Zn in ppm and (d) C_org in (%) are represented for the 1980s. A larger number of elements could be retrieved in this decade, but research was limited to the Inner Saronikos Gulf. It can be noted that the maximum C_org and Cr values are detected in the strait between Psyttalia and Keratsini and are probably due to ASO operations, whilst for the rest of the represented elements, the maximum values are located closer mainly to the Fertilizer Plant and Piraeus Port. Co and Fe present similar patterns to that of Zn (Figure S1_1a,b, Supplementary File_S1). Figure 9 depicts the surficial distributions of (a) As, (b) Cu, and (c) Zn in ppm and (d) C_org in (%) during the 1990s. The main characteristic of this decade is the sampling area expansion into the entire Saronikos Gulf. The results revealed another trend in the history of heavy metal pollution in the gulf, which is the increased metal concentration along the eastern shore of the Attica peninsula and another one in western Elefsina Bay. The elevated values of Co, V, and Cr (Figure S2_1a,b,d Supplementary File_S1) in the western Saronikos Gulf are attributed to natural inputs from ultrabasaltic rocks [26]. More particularly, the distribution patterns of As (Figure 9a) and Pb (Figure S2_1c Supplementary File_S1) are similar to the maximum values in the Salamina straits and the elevated values along the southeast of Perama. Additionally, Cu and Zn (Figure 9b,c) present similar dispersions in their maximum values to those of As (Figure 9a) and Pb (Figure S2_1c Supplementary File_S1), although the Cu values are elevated in Elefsina Bay, and the elevated Zn values do not disperse at the Attica shoreline east of Piraeus (Figure 9b,c). In this decade, the C_org (Figure 9d) distribution presented two maxima: the first one in the Salamina straits complying with As, Pb, Cu, and Zn, and another one north at Piraeus. Finally, V shows a different pattern, with maximum values in Elefsina Bay (Figure S2_1b Supplementary File_S1). Subsequently, for the 2000s, distribution maps were produced (Figure 10) for (a) As, (b) Cu, and (c) Zn in ppm and for (d) C_org in (%), and for Cr, Ni, Pb, and V in ppm (presented in Figure S3_1a, b, c, and d, respectively, in Supplementary File_S1). Based on the available data, marine sediment pollution affects the Inner Saronikos Gulf, Elefsina Bay, and Piraeus Port, as well as the southern coastline of Salamina Island, with Cr, Ni, and V affecting a larger area of the gulf. Thus, Cr, Ni, and V seem to follow a similar pattern (Figure S3_1a,b,d Supplementary File_S1), while Cu, Zn, Pb, As, and C_org present similar distributions. Finally, Figure 11a–d depict the As, Cu, and Zn (in ppm) and C_org (%) surficial distributions in the 2010s, whilst Figure S4_1 (Supplementary File_S1) presents the (a) Cr, (b) Co, (c) Pb, (d) Ni, and (e) V distributions. In this decade, sampling points are spread throughout the entire gulf, though there are less data on C_org. It can be observed that the same metal grouping as that of the previous decade is pictured. There is one group of elements (Cu, As, Zn, Pb, C_org) with elevated values only at Elefsina Bay and the Inner Saronikos Gulf, representing mainly the anthropogenic influence.

Additionally, there is a second one (Co, Cr, Ni, V) with high values at Elefsina Bay and the western Saronikos Gulf reflecting human contamination in the first region and geological conditions in the latter one. Moreover, it could be noted that V has presented mainly elevated values at Elefsina Bay since the 1990s. At this point, it should be mentioned that, in some cases, differences in the distribution patterns of metals over the decades could be attributed to a lack of available data. Thus, it is difficult to extract certain conclusions for the metals’ spatial and temporal distributions.

Considering the abovementioned, it can be concluded that the Inner Saronikos Gulf, Elefsina Bay, and part of the eastern coastline of the Saronikos Gulf have mainly been affected by human contamination consistently since the 1980s. The elevated metal values in the western Saronikos basin can be attributed to natural sources. Furthermore, in recent research programs, more metals are being investigated, and sample networks are growing. This is important because systematic monitoring, data exchange and availability, and well-defined environmental regulations are required.

3.2.2. Environmental Indices

Pollution Loading Index (PLI).

The calculation of the Pollution Loading Index offers an estimation of the environmental burden due to the overall multi-metal pollution and was estimated per decade. The metals used differ per decade depending on the data available as well as the sediment sampling design. Thus, in the 1970s, the PLI could not be estimated due to the lack of a dataset with metal concentrations.

In the 1980s, the PLI was calculated based on the metals Co, Cr, and Zn and presented values between 1.7 and 13.3 (

Table 7. PLI values compared to the number of samples and extent of area sampling per decade in the Saronikos Gulf.

Decade	No. of Samples	Extent of Area of Sampling (km2)	PLI Values	Overall PLI Area
1970s	30	224	-	-
1980s	28	29.5	1.7–13.3	4.8
1990s	62	1921	0.7–43.2	14.7
2000s	57	436	0.5–6.4	2.7
2010s	80	2277	0.6–4.2	1.9

), whilst the overall PLI area was also high (PLI = 4.8) and comparable to polluted regions worldwide.

In the 1990s, the Pollution Loading Index was calculated based on the metals Mn, Ni, Zn, Fe, Cr, Cu, Pb, Mg, V, Co, and As, since more metals were studied in this decade and a larger area of the Saronikos Gulf had been sampled. Values for the PLI were estimated between 0.7 and 43.2 (Table 7), with the high values located in the Keratsini harbor [43]. Calculating the overall PLI area, the index is higher than that in the previous decades, indicating the continuous pollution burden due to heavy metal inputs being introduced in the coastal environment (PLI = 14.7).

The PLI area was calculated as 2.7 for the 2000s, based on the metals V, Cr, Mn, Co, Ni, Cu, Zn, As, and Pb, with PLI values ranging from 0.5 to 6.4 (Table 7). The higher values were estimated for sampling sites from Elefsina Bay and the Inner Saronikos Gulf (Salamina–Perama strait). Moreover, the PLI area was also calculated for the next decade, 2010, as follows: PLI area = 1.9, based on the same metals (V, Cr, Mn, Co, Ni, Cu, Zn, As, Pb) with a range of zone values between 0.6 and 4.2. Figure 12 compares PLI zones (categories grouping samples with similar PLI values) for the 1990s, 2000s, and 2010s, and as can be seen, the overall pollution seems to have declined in the most recent decades but remains high though for most of the sampling sites. According to Figure 12, PLI zones 1 to 4 present similar values for the three decades. On the contrary, in PLI zones 5 to 8, the PLI values for the 1990s are extremely high compared to those of the other two decades. This is attributed to the very high metal concentrations estimated in sediment samples collected from Keratsini Bay in this decade. For simplicity reasons, zone 8 in the chart (Figure 12) represents PLI values between 18 and 43.

Figure 13a–d depict the spatial distribution of the calculated PLI values for each sampling site for the 1980s, 1990s, 2000s, and 2010s. As can be seen, apart from the fact that the seriously affected area extends to the Inner Saronikos (Salamina straits) and Elefsina Bay, the PLI was not calculated below 1 for any sampling site of the entire area, with the exception of the south Attica coastal zone. Based on the PLI spatial distribution in the 1980s (Figure 13a) and 1990s (Figure 13b), extremely polluted sediments were located at Piraeus Port. In the next decade, apart from the Salamina straits, elevated PLI values were also observed at Elefsina Βay (Figure 13c). In the 2010s, the PLI was calculated for the surficial sediments of the entire Saronikos Gulf, presenting elevated values mainly at Elefsina Bay and lower values (but still > 1) for the Salamina straits.

Additionally, in Figure 14, the subareas of the Saronikos Gulf are highlighted where the PLI values held their maximum levels in every decade according to the available data. As can be seen, the most affected sites include the Piraeus Port, the Salamina straits, Elefsina Bay, and the island of Psyttalia. It seems that the highly impacted area expanded from the earlier to the most recent decades, and that the Salamina straits with the west Attica coast were the most affected during the four decades.

The comparison of the overall PLI area by decade is a challenging goal since the sampling design led to the spatial distribution of the sediment samples, which are not homogeneously distributed in the gulf and, in some cases, are not completely representative. For example, the sampling sites in the 1990s, based on which the PLI of this decade was calculated, are mainly concentrated in the highly polluted port of Keratsini, leading to very high PLI values. In the following decades (2000s and 2010s), the port of Keratsini was represented by very few or no samples, resulting in a decrease in the PLI value. For the last two decades, the 2000s and 2010s, where the dispersion of sampling sites is comparable, a decreasing trend in the PLI is observed.

3.2.3. Potential Biological Effects of Heavy Metals

Potential Ecological Risk Index (PERI).

The Potential Ecological Risk Index was calculated using the average concentrations of the metals As, Cr, Cu, Pb, and Zn according to Hakanson [70] per decade. In the 1970s, the PERI was not able to be estimated. In the 1980s, the Potential Ecological Risk Index was calculated based on As, Cr, and Zn, and sediments were characterized by a «moderate ecological risk», with a mean RI of 228, (150 < RI < 300). In the 1990s, the PERI was calculated using the metals Cr, Cu, Pb, and Zn, and Saronikos Gulf surficial sediments were exposed to «considerable ecological risk», with an RI of 309. Due to the data availability over the following two decades, the metals used to calculate the Potential Ecological Risk Index were As, Cr, Cu, Pb, and Zn. In the 2000s, the surficial sediments with mean RI values of 124 (RI < 150) were exposed to «low ecological risk», as well as in the 2010s, with an RI of 84. Figure 15a,b depict the mean PERI values for the 1990s, 2000s, and 2010s using Cr, Cu, Zn, As, and Pb. In Figure 15a, the PERI values for each metal are compared for the decades, demonstrating higher values for Pb and Cu (from the 1990s and 2000s). Finally, Figure 15b illustrates the decreasing trend of the PERI from the 1990s to the most recent decade.

Human Health Risk Assessment (HHRI).

The results of the Human Health Risk Assessment of heavy metals (mean values of the HQi and HI for children and adults) calculated for Saronikos Gulf seafloor sediments are presented in Table S3_1 Supplementary File_S1. In the 1970s and 1980s, health risks were calculated for Zn, and the values did not exceed 1, which is an indication that there were no potential non-carcinogenic health risks from this metal at the time. In the 1990s, the values of the HQi and HI for children and adults were calculated for Zn, Cu, and Pb. As can be seen in Table S3_1 Supplementary File_S1, the HI mean values for children exceeded 1 (HI > 1), indicating that children had more chances of non-carcinogenic risks than adults. Additionally, in the 1990s, the HI > 1 and Hqi > 1 values were calculated, attributed to Cd and Pb, indicating potential non-carcinogenic risks due to these metals as well for children and adults. It should be noted that the samples indicating potential health risks due to Cd and Pb were collected at the Keratsini harbor. In the 2000s and 2010s, the health risks were calculated due to Cu, Zn, and Pb, and their mean values (HI < 1) indicate no potential health risks, but as is depicted in Figure 16a,b and Figure 17a,b, the HQ_ing and HI values for Pb (in the 2000s) exceed 1 in samples derived from the Inner Saronikos Gulf (stations were selected from data obtained by Karageorgis et al. [26,90,106] and Gkaragkouni et al. [27]). Thus, there were chances of non-carcinogenic risks for both children and adults due to Pb in the 2000s in certain subareas of the gulf (Elefsina Bay, the Inner Saronikos Gulf, and the Psyttalia–Keratsini strait). In general, the HI values of heavy metals for children and adults decreased in the order of Cd > Pb > Cu > Zn in all decades. When comparing the HI values for Zn, which is a commonly studied element (over the past 50 years), the following ranking was formed according to the HI values: 1990 > 1970 > 1980 > 2000 > 2010. Therefore, comparing the HI values in the Saronikos Gulf, a decreasing trend has been documented in the most recently deposited sediments.

4. Conclusions and Recommendations

The Saronikos Gulf has received the attention of researchers over the last fifty years, as it is considered the most polluted marine area in Greece. In this period, the heavy metal pollution surveys in the gulf were almost equally based upon surface sediment and sediment core sampling. Most studies focused their interest on specific subareas of the gulf at the proximity of pollution point and non-point sources. These areas are the Salamina–Perama strait, around the Psyttalia Wastewater Treatment Plant and Athens Sewage Outfall, and to a lesser extent, Elefsina Bay.

The number of publications (in %) per decade reveals the 1980s as the most productive decade for the topic of heavy metal pollution in the Saronikos Gulf, followed by the 2000s and 2010s. It can be also concluded that the researchers’ interest between 1974 and 2024 was considered approximately steady, without great fluctuations. Among the heavy metals, Zn has been the most studied in the 50 years of research in the Saronikos Gulf, followed by Pb, Co, and Cr. The organic carbon content is a crucial element for pollution studies, making it an essential part of elemental analysis datasets. In contrast, Hg was only studied in the 1970s and could be an element requiring further investigation, especially because its toxicity is highly dependent on its inorganic/organic speciation. The same applies to V and U, as more evidence should be gathered for their inputs and paths in the marine environment of the Saronikos Gulf. To date, very few studies on heavy metal speciation have been conducted, not allowing for the in-depth study of the geochemical behavior of critical heavy metals. Additionally, very few research papers present REE values, natural radionuclide concentrations, and only single-element studies.

The introduction of new, sophisticated analysis tools and techniques can significantly aid in deepening our understanding of the heavy metal contamination in various areas. Making a timeline for the instruments used in heavy metal analysis in the gulf, the following ranking can be noted: INNA-AAS-XRF and ICP-MS. Portable and smart devices or even remote sensing could be helpful according to Lu et al. [107] and Wang et al. [108], but in any case, developments in analysis, instruments, and methods are expected to be adapted to coastal environments and to help monitor and protect some of the most valuable but anthropogenically impacted ecosystems [3].

Estimating and defining the background (preindustrial) metal concentrations is a crucial point, since all the environmental indices are based on the ratio of a metal’s concentration to its background value. Background values were estimated in 12 of the research papers studied for Saronikos Gulf sediments. In a few of them, unpolluted surficial sediments, far away from the pollution sources, were used. The use of sediment cores is the most reliable, as this approach not only provides background levels but also produces a record of contamination and, if dated, establishes the pollution “history” of the area and the possible relaxation rate. Although the sedimentation rate varies significantly depending on the subregion of the gulf, based on the available data, the rate ranges from 0.006 cm/year to 1.0 cm/year, with a mean value of 0.41 cm/year. Based on this mean value, the uppermost 25–30 cm of sediments can be considered contaminated, taking into consideration the initiation of industrialization in the gulf. Additionally, as core samples give an undisturbed cross section beneath the seabed and reveal temporal heavy metal variations, there is the need to obtain more sediment core data.

Multivariate statistical methods (PCA, factor analysis) have the advantage of reducing the dataset and extracting a small number of statistically dominant factors that represent geochemical phases that can lead to the identification of metal pollution sources. Determining the sources of trace metals in a relatively small area of the Saronikos Gulf, such as the Salamis straits and the surrounding area, which are influenced by different polluting activities, is a very challenging task. Factor analyses of different datasets from different decades showed three main groups of heavy metals related to specific polluting sources, if these could allocate the metals to specific sources. The first group consists of As, Ag, Cd, Cr, Cu, Zn, Pb, and C_org and is attributed to the ASO, WWTP, and possibly the FP. This group is dominant at the Salamina–Perama strait and the surrounding area of Psyttalia Island (WWTP). The second group of metals consists of Fe, Mn, Zn, Pb, Cu, Mo, As, Mg, and TOC and is related to the steel industries located at Elefsina Bay. The third pollution factor was appointed to crude oil refineries, introducing mainly Sn and V into the gulf.

Following the above and based on the available environmental indices calculated for the gulf, the metals Pb, Cd, As, Cr, Hg, Sb, Zn, Se, Cu, Mo, Ag, Sn, and Br show high contamination levels in the Inner Saronikos Gulf, Elefsina Bay, and the Salamina–Perama strait. Moreover, the PLI site values provided a spatial expression of the gulf’s pollution by heavy metals per decade. The highest values were found in the 1990s in a limited area (Keratsini harbor), without these results being completely comparable with those of the two following decades due to different sampling designs. In the last two decades, a decrease in the PLI was found due to the operation of the WWTP and probably the reduction in industrial operations. It should be emphasized that different sediment sampling designs have a strong impact on the comparability of results, resulting in the insufficient monitoring of pollution.

The Potential Ecological Risk Index and Human Health Risk Assessment indicate Pb as a metal of high ecological risk and the most likely to expose children and adults to non-carcinogenic risks. Thus, further research must be undertaken in the direction of ecological risks and Human Health Risk Assessment. The carcinogenic risk due to heavy metal exposure could also be an aspect of inspection in the area.

Since coastland regions are usually affected by multiple anthropogenic stressors, it is essential to examine the combined impacts of heavy metal pollution with other anthropogenic environmental changes, in particular climate change, organic pollutants, and eutrophication, rather than heavy metal pollution alone [109]. Multi-factorial studies are required to uncover the effects of heavy metals in more realistic circumstances because heavy metal pollution and other stressors may frequently interact in different ways [3].

Although more research is required to better understand the dynamics of pollutants regarding coastal and shallow marine sedimentary processes and reduce the degree of pollution, the results of this study can aid in improving the area’s environmental management strategy. In addition, to determine the historical evolution of the pollution, collecting sediment cores in designated areas selected based on the results of the present research is strongly suggested. To completely comprehend the processes of metal accumulation, it is therefore essential to continuously monitor the sources of heavy metals and conduct additional research on the sedimentary records. This calls for careful monitoring studies to assess the current and future levels of heavy metal contamination and to recommend possible preventative and precautionary actions. A continuous and robust monitoring and pollution control plan should be implemented in areas severely affected by heavy metal inputs for the sustainable development of the coastal economy and environment.