AB STRA CT Geochemical analysis and trace element distribution were performed on the marine sediments from short cores (30 to 50 cm) from the Middle (Jabuka and Palagruža pits, depth 230 m and 170 m, respectively) and South Adriatic Sea (depth 1030 m) and the Albanian offshore (50 m depth). The distribution of trace elements (As, Co, Cd, Cu Cr, Ni, Zn, Hg, and Pb) and major elements (Al, Ca and Mn) and mineralogy in the sediments is presented. Sediments are highly heterogeneous and consist of carbonate and detrital aluminosilicate minerals. The main mineral phase is cal- cite, followed by quartz, feldspars, micas and clay minerals (smectite, chlorite, illite and kaolinite). The cores were dated using 137Cs. The cores from the South Adriatic Pit and Palagruža Sill gave estimated sedimentation rates of 1.8 mm y–1 and 3.1 mm y–1 in Jabuka Pit. Distributions of Ni and Cr showed that they can be used as tracers of sedi- ment provenance along the southern part of the Eastern Adriatic Current. Calculated enrichment factors for Pb, Cd and Hg are highest in the top 2 cm of the cores. Mercury shows the highest degree of enrichment in 0–2 cm sediment intervals (the highest in the Albania core). Generally the estimated surface enrichment follows the order: Hg>Pb>Cd. No enrichment was found for Zn, Ni and Cr. Keywords: trace metals, accumulation rate, marine sediment record, enrichment factors, Adriatic Sea, Eastern Adriatic Current Geologia Croatica 67/3 185–205 10 Figs. 3 Tabs. 3 Supp. Tabs. Zagreb 2014 Metal deposition in deep sediments from the Central and South Adriatic Sea  Nikolina Ilijanić1, Slobodan Miko1, Branko Petrinec2 and Zdenko Franić2 1 Croatian Geological Survey, Sachsova 2, HR-10000 Zagreb, Croatia; (corresponding author: nilijanic@hgi-cgs.hr) 2 Institute for Medical Research and Occupational Health, Radiation Protection Unit, Ksaverska cesta 2, HR-10001 Zagreb, Croatia doi: 10.4154/gc.2014.14 marine and estuarine environments along the eastern Adri- atic coast (PROHIĆ & KNIEWALD, 1987; BOGNER et al., 1998; MIKO et al., 2007; SONDI et al., 2008; CUCULIĆ et al., 2009; CUKROV et al., 2011, OBHODAŠ et al., 2012; LOVRENČIĆ MIKELIĆ et al., 2013). Trace metal data are also available from relatively few long cores spanning through the Holocene and into the Pleistocene (CALANCHI et al., 1996; LUCCHINI et al., 2003) in Mid Adriatic De- pression (MAD), and basin-wide low sampling density geo- chemical studies of surface sediments (DOLENEC et al., 1998; DE LAZZARI et al., 2004). SPAGNOLI et al. (2010) gave a detailed overview of the studies of trace elements in sediments of the central-southern Adriatic Sea, concluding that data indicating contamination are only limited to spe- cifi c areas. Based on the analysis of a core from the MAD, CALANCHI et al. (1996) concluded that it is consistent with 1. INTRODUCTION Down-core and surface geochemical and mineralogical ana- lysis have been successfully used as tools in reconstructing past variations in sedimentation, provenance and pollution studies of the Adriatic Sea. The trace element geochemistry studies of the Adriatic basin can be divided into several groups, the pollution and provenance studies related to the Western Adriatic Current (WAC) along the eastern coast of Italy (DINELLI & LUCCHINI, 1999; AMOROSI et al., 2002; DINELLI et al., 2007; SPAGNOLI et al., 2008; AN- NIBALDI et al., 2009; AMOROSI, 2012; GOUDEAU et al., 2013; ROMANO et al., 2013), the pollution related to the studies of the Gulf of Trieste (COVELLI et al., 2001, 2006, 2007; FAGANELI et al., 2003; COVELLI et al., 2006, 2007; ACQUAVITA et al., 2012; EMILI et al., 2012), the coastal Geologia CroaticaGeologia Croatica Geologia Croatica 67/3Geologia Croatica 186 the average composition of fine-grained Middle Adriatic se- di ments, and that major variations in trace element concen- trations depend on changes in the silicate/carbonate ratios, and the type and abundance of silicate supply. The study by TANKÈRÉ et al. (2000) used a simple box mass balance mo del for Mn, Fe, Pb, Zn, Cu, Ni on an annual scale for the different compartments of the Adriatic Sea. The model was based on fluxes along the WAC. They concluded that the con tribution of the Northern Adriatic to the rest of the Adri- atic Sea was relatively small because of the restricted water exchange and the loss due to burial in the sediment. On an annual scale, the Adriatic Sea appeared to be a source of dis- solved Cu, Mn and Fe for the Mediterranean Sea through the Strait of Otranto whereas for dissolved Zn and Pb the Adri- atic Sea appeared to be a net sink. For dissolved Ni, inputs and outputs through the Strait of Otranto balanced each other. TOMADIN (2000) used clay mineral composition and distribution in the central and southern Adriatic Sea, as tra- cers of sediment provenance, indicating longitudinal disper- sion connected with the general Adriatic cyclonic circulation. He recognized fluxes directed south-westward along the Ital- ian offshore – the Apennine flux (Western Adriatic Current- WAC) near the coast and Padane flux in the open sea. The NW flux flows along the eastern coast (Eastern Adriatic Cur- rent–EAC, GIANI et al., 2012), of the central and southern Adriatic basin, which he termed as the “Albanian flux”. Studies on the geochemistry of the core samples have been undertaken for the western part of the Adriatic Sea, us- ing the geochemistry as a tracer for both provenance and pollution studies along the WAC (AMOROSI et al., 2002; DINELLI et al.; 2007; WELTJE & BROMMER, 2011; GOU DEAU et al., 2013; ROMANO et al., 2013). Only the studies of DOLENEC et al. (1998) and DE LAZZARI et al. (2004) provided geochemical data for surface sediments along the southern part of the Eastern Adriatic Current (EAC). DO- LENEC et al. (1998) indicated that the main sources of trace element accumulation in the central Adriatic were both the WAC and EAC. The EAC was responsible for the accumu- lation of As, Co, Cr, Cu and Ni, V, and Zn with a contribu- tion of WAC to Cu, V and Zn. The sources of Hg were con- sidered to be the northern Adriatic and the Apennine mainland (DOLENEC et al., 1998). The geochemistry and mineralogy of Mn from the central Adriatic in surface sediments was also given by DOLENEC et al. (1998). The Albanian River inputs are considered to be major sources of Co, Cr, Ni, V and Zn in the eastern part of the southern Adriatic (DO- LENEC et al., 1998) as a consequence of weathering and erosion of catchments containing ophiolitic mélanges asso- ciated with ophiolitic sequences in the Mirdita–Subpelago- nian zone. The high concentrations of transition metals (such as Ni, Co, Cr and V) are one of their primary geochemical features (SACCANI & PHOTIADES, 2005). This study reports the depth profiles of the eight trace elements usually identified as priority contaminants in aqua- tic systems As, Cd, Cr, Cu, Hg, Ni, Pb and Zn as well as Mo, Mn, Al and Ca as geogenic elements in 5 short marine sedi- ment cores, collected at locations on a transect along the western rim of the EAC. Sediment cores provided evidence for the accumulation of metals in the deep sea environments of the Adriatic. This study provides an insight to the possible anthropogenic influences in the area. To assess the distribu- tion and the degree of sediment contamination and to distin- guish natural and anthropogenic inputs, metal-enrichment factors (EF) were calculated (FÖRSTNER & WITTMANN, 1981; LI, 1981). Mineralogical analysis was performed in order to determine the nature and their mutual comparabil- ity as well as down-core variations of the analyzed cores. Based on the metal concentrations measured in both surface and deeper sediments, trace element enrichment factors were also calculated. The depth profiles represent the signature of historical contamination. Radioactive isotopes (137Cs) cou- pled with variations in elemental concentrations allowed re- construction of the history of trace element accumulations in Adriatic sediments through time. 2. STUDY AREA The Adriatic Sea is a narrow and shallow epicontinental ba- sin (approximately 200×800 km), forming a distinct sub-re- gion within the Mediterranean Sea (Figure 1). The surface area of the whole Adriatic Sea is 138,595 km2 (BULJAN & ZORE-ARMANDA, 1976). The Adriatic Sea may be divided into three parts (Figure 1): (1) the Northern Adriatic Sea from the Trieste gulf to the Ancona promontory and the island of Pag; (2) the Central Adriatic Sea from the Ancona promontory and the island of Pag up to the transect Gargano-Split; and (3) the Southern Adriatic Sea from the Gargano-Split to Otranto. The central Adriatic Sea again lies on the continental shelf, but is char- acterized by raised areas, which sometimes emerge to pro- duce isles, and by a large depression up to 272 m deep. The Southern Adriatic Sea is very different and can be divided into three units: the continental shelf, the continental slope and a bathyal plain at 1233 m. The seabed is covered with sand and in some places with a mixture of sand, mud and silt. The northern part is sandy, while the middle and south- ern parts are dominantly covered in silt (PIGORINI, 1968; LEDER, 2004; SPAGNOLI et al., 2008, SPAGNOLI et al., 2010). The Adriatic Sea is located between two major climatic regimes. The northern part is highly influenced by the tem- perate and humid climate of southeast Europe, while the southern part is governed by the arid conditions of the Med- iterranean and northern Africa. The Adriatic Sea has a mic- rotidal regime and is dominated by a cyclonic circulation driven by thermohaline currents (ORLIĆ et al., 1992). The surface water of the Adriatic Sea circulates in an anticlock- wise direction. It flows parallel to the eastern coast to the north (NW direction, EAC; GIANI et al., 2012), then turns towards the western part on the Italian coastline to the south (Western Adriatic Current – WAC). Global cyclonic circula- tion is divided into three re-circulation cells (Figure 1) in the northern, central and southern sub-basins, induced by the strong Bora wind and being controlled by the bathymetry of the Jabuka and South Adriatic Pits (ZORE ARMANDA & GAČIĆ, 1987; ORLIĆ et al., 1992). Ilijanić et al.: Metal deposition in deep sediments from the Central and South Adriatic Sea Geologia Croatica 187 3. MATERIALS AND METHODS In this study 5 box-cores were analysed, from the central and southern part of the Adriatic sea (Figure 1, Table 1). One core was from the Mid Adriatic Depression (“Jabuka pit”, the core was termed as Jabuka), while a second came from the Palagruža Sill (“Palagruža” core). Two cores are from the South Adriatic Pit (cores SA PIT-1 and 2) and an offshore core came from the Albanian coast (“Albania core”). The sediment cores were collected by means of a box-corer, which provided virtually undisturbed samples up to 50 cm long. The cores were cut into 2 cm intervals, freeze-dried after coring on the boat and transported to the laboratory. Samples were powdered manually in an agate mortar and Figure 1: Map showing the Adriatic Sea and locations from which sediment cores were taken. Global cyclonic circulation (arrows) of the marine currents in the Adriatic Sea is broken into three re-circulation cells in the northern, central and southern sub-basins, being controlled by the bathymetry of the Mid Adriatic Depression and South Adriatic Pits. WAC=Western Adriatic Current; EAC=Eastern Adriatic Current. Table 1: List of the samples, coordinates and depths of the locations. Location Geographic coordinates Depth (m) Core length Jabuka N 43˚03’10.25’’ E 15˚16’30.94’’ 230.8 50 cm Palagruža N 42˚28’34.86’’ E 16˚11’27.09’’ 169.8 40 cm SA PIT 1 N 42˚20’20.77’’ E 17˚47’18.21’’ 1041.4 42 cm SA PIT 2 N 42˚20’44.70’’ E 17˚44’49.09’’ 1030.0 42 cm Albania N 41˚43’24.33’’ E 19˚19’54.29’’ 59.0 38 cm Geologia Croatica 67/3Geologia Croatica 188 analysed for bulk mineral and geochemical composition and total mercury concentrations. The mineral composition of all samples was determined by a PANalytical X’Pert Powder X-ray diffractometer, equipped with Ni-filter CuKα radiation, vertical goniometer with θ/θ geometry and a PIXcel detector. Scan conditions were: 45kV and 40 mA, ¼ divergence slit and antiscatter slits, step size 0.02° 2θ, time per step 2s, in the range of 5–65° 2θ. After grinding, samples were sieved through a 0.4 mm sieve and ground in McCrone micronizing mill to reduce material to <5 μm powder samples and ensure random min- eral orientation. This preparation method facilitated semi- quantitative analysis, according to EBERL (2003). In addi- tion, oriented samples were prepared for clay mineral ana ly sis. Calcite was removed by treating the sample with buffered sodium acetate (NaOAc) solution at pH 5.0 in a 1:10 sedi- ment/buffer ratio. The pH of the slurry was lowered to 5.0 using HCl. The <2μm fraction was separated from selected samples by centrifugation (1 minute, 2000 rpm) to obtain mineralogical details about the clay-size fraction and clay minerals. Oriented mounts were prepared on glass slides with an eye-dropper. These samples were analyzed in air-dried, ethylene glycol-solvated, and heated (400°C and 550°C) states and scanned in the range of 2–30° 2θ. The XRD pat- terns were interpreted following MOORE & REYNOLDS (1997). Quantitative mineral analysis was undertaken using the RockJock, computer program for QXRD (RkJock.xls; EBERL, 2003). The wt% of minerals in a sample was cal- culated from integrated X-ray intensities. The RockJock technique has been checked for accuracy using artificial mix- tures and generally produced answers that are within ± 2 to 5 % of actual values (EBERL, 2003). Chemical analysis was conducted on all 41 samples. The dissolution of 0.25 g of sediment was performed with a mix- ture of HCl–HF–HClO4–HNO3 acids on a hot plate until dry. The residue was then treated with diluted HCl–HNO3–H2O (15% aqua regia) in a hot bath (> 95 °C) for 30 min. The re- sulting solutions were analyzed by mass spectroscopy using a Perkin Elmer Elan 6000 or 9000 ICP-MS for a set of 41 elements: Ag, Al, As, Au, Ba, Be, Bi, Ca, Cd, Ce, Co, Cr, Cu, Fe, Hf, K, La, Li, Mg, Mn, Mo, Na, Nb, Ni, P, Pb, Rb, S, Sb, Sc, Sn, Sr, Ta, Th, Ti, U, V, W, Y, Zn, and Zr. Total mercury concentration was determined after HCl:HNO3 (3:1, aqua regia) extraction of subsamples by cold vapour AAS. From this set of elements potential contaminants including As, Cd, Cr, Cu, Ni, Hg, Pb and Zn were selected to evaluate metal/contamination in the cores, as well as Mo, Mn, Al and Ca as the presumed natural variables. The chemical analysis was performed at the ACME Analytical Laboratories in Van- couver, Canada. The accuracy was checked with certified reference materials (GXR-2, GXR 5, and SJS-1; USGS). The accuracy for most elements analysed in reference materials was <10 % of the certified values. Precision was determined by repeated analyses of both certified reference samples and randomly selected sediment samples (every 5th sample in the batch) with the resulting average coefficient of variation of approximately 5%. The five sediment cores were measured for 137Cs activ- ities in order to derive sediment accumulation rates. Details of the performed analysis and calculations of accumulation rates are given in PETRINEC et al. (2012). Descriptive data analyses (mean, standard deviation, maximum and minimum concentrations) and calculation of enrichment factors (FERGUSON, 1990; RIDGWAY & SHIM MIELD, 2002) were carried out. In addition, correla- tion factors were calculated to determine and the island of Pag up to the transect Gargano-Split; asses the processes in- volved. The data sets for the analyzed cores (this study) and the data from DOLENEC et al. (1998) were analyzed using the STATISTICA 7 (StatSoft, 2006) statistical program. The comparison of geochemical data available for Adriatic ma- rine sediments given by DOLENEC et al. (1998) and GOU- DEAU et al. (2013) was possible since the same methods based on dissolution with the mixture of HCl–HF–HClO4– HNO3 for determination of element contents were used in their studies. 4. RESULTS 4.1. Sediment characteristics Mineral composition of analysed sediment samples is shown in Figure 2, and Table 2. The predominant mineral phase in all samples was calcite (14–34%), the highest values were in Palagruža (28–34%) and the lowest in the Albania samples (10–15%). Mg-calcite was present in all samples in concen- Figure 2: Mineral composition of the representative samples from the five locations. Symbols: Qtz-quartz, Cal-calcite, Kfs-potassium feldspar, Plg-pla- gioclase, Ms/I-muscovite/illite, Sm-smectite, Chl-chlorite, Kln-kaolinite, Px- pyroxene, Amph-amphibole, Mg-cal-Mg calcite, Hl-halite. Ilijanić et al.: Metal deposition in deep sediments from the Central and South Adriatic Sea Geologia Croatica 189 trations from 1–7%, and seems to be higher in Palagruža (up to 10%). Quartz occurs in amounts from 3–15% and it was higher in samples from Albania (10–15%) and lower in Palagruža (2–10%), (Figure 3). Feldspars are also abundant, and potassium feldspars dominate over plagioclase. Trace amounts of pyroxene and amphibole appeared in some sam- ples. Clay minerals are represented by smectite, illite, chlorite and kaolinite (Figure 4). Smectite is the dominant clay min- eral in all the analysed sediment cores (15–43%), but with a slightly lower presence in the Albania cores (18–27%). Illite and chlorite have the highest contents in sediments from the Albania core (7–18%, 11–16%, respecti vely), while illite was depleted in the Palagruža core (3–9%). Kaolinite is present in all samples, and the content ranging from 1–9%. 4.2. Accumulation rates The highest sedimentation rates have been found in the cores from Albania (~4 mm/y) and Jabuka (3.1 mm/y), and 1.8 mm/y in Palagruža. In the south Adriatic, where two peaks are more easily identified, sedimentation rate could be esti- mated to be approximately 1.8 ± 0.5 mm/y which is quite similar to the value estimated for Palagruža. Results indicate that for the Palagruža Sill and the South Adriatic Pit, an an- nual average sedimentation rate of 1.8 ± 0.5 mm is likely and that the top 10 cm interval of the sediment core represents approximately 50 years of history and a 40 cm core could be an equivalent to 200 years. The Jabuka and Albania cores therefore span much shorter histories (approximately 130 and 100 years, respectively). However, it should be noted that these results are only approximate due to the method limitations and large uncertainties in the estimation of the exact location of 137Cs activity concentration peaks in the sediment profiles. 4.3. Variations of elemental concentrations with space and depth The tables containing geochemical data are given as online supplementary tables. The concentrations of analyzed major and trace elements are summarized in Supplementary Table 1. The down-core contents of major elements Ca and Al dis- Table 2: Quantitative mineral analysis of the samples (SA PIT 1, SA PIT 2, Jabuka, Palagruža, Albania) (RockJock; EBERL, 2003). Results are in wt% and rep- resent average data of the samples in the 10 centimetre interval. JABUKA PALAGRUŽA SA PIT 1 Minerals 0–10 10–20 20–30 0–10 10–20 20–30 0–10 10–20 20–30 Quartz 8 7 6 9 12 8 9 7 8 Calcite 23 29 30 22 12 27 23 24 24 Mg-calcite 4 4 5 5 2 2 4 4 4 K-feldspar 6 8 9 5 8 7 7 7 6 Plagioclase 5 6 7 5 8 6 7 4 5 Illite 5 5 6 10 11 7 10 9 10 Smectite 33 23 23 26 24 24 37 26 25 Chlorite 9 7 7 8 14 9 12 8 9 Kaolinite 4 6 5 6 6 5 5 6 6 Pyroxene 1 1 1 1 – 1 – 1 1 Amphibole 1 3 2 2 2 3 3 1 2 Halite 1 2 2 2 2 2 2 2 1 SA PIT 2 ALBANIA Minerals 0–10 10–20 20–30 0–10 10–20 20–30 Quartz 13 13 6 9 9 7 Calcite 13 13 20 21 23 20 Mg-calcite – 1 3 4 2 2 K-feldspar 7 7 7 7 6 6 Plagioclase 9 8 5 5 6 6 Illite 14 15 6 11 8 6 Smectite 21 21 32 24 29 34 Chlorite 13 12 10 9 9 11 Kaolinite 7 7 5 6 5 4 Pyroxene – – 3 – – 2 Amphibole 2 2 2 2 3 2 Halite 1 1 1 2 2 2 Geologia Croatica 67/3Geologia Croatica 190 play narrow ranges in the individual cores. The Ca concen- trations are highest in the Palagruža core (13.68–16.81%) and the lowest in the Albania core (4.83–5.19%). Several samples that have higher Ca contents and fall outside the Ca- calcite line indicate Ca to be linked with some aluminosili- cate phases (smectite, chlorite). Relationships between ma- jor elements Ca, Al, Ti and calcite and muscovite/illite in Figure 5 clearly display these relationships. Most samples in the cores contain on average < 30% smectite and < 10% chlorite with the exception of the Albania core where the contents of chlorite range from 12–14% (Table 2). Such a deviation in Ca content that is not related to carbonate con- tent could be attributed to the presence of Ca-rich plagio- clase, as a signature of a high detrital siliciclastic input. The cores from the Mid Adriatic Depression (Jabuka) and South Adriatic Pit (SA PIT 1&2) have similar Ca content (11.1% and 10.7%, respectively) as well as calcite (20–24%). Al and Ti contents are indicative of detrital delivery and the highest contents are in the Albania offshore cores. Aver- age Al content ranged from 6.6% in the Albania core to 4.9% in the Palagruža core. The cores from the depressions had average Al contents from 5.7 % (Jabuka) to 6.1% (SA PIT). Ti contents were similar in the Jabuka, SA PIT and Palagruža cores (Supplementary Table 1), while it was higher in the Albanian core (0.34 to 0.37%). These elements and their re- lationship to trace elements are important in evaluating pos- sible pollution. The trace element composition of the analyzed cores shows higher spatial and down-core variability (Supplemen- tary Table 1). The down-core concentration profiles of Pb, Hg, Zn, Cd, Cr, Ni, Cu, As, Mo Mn and Zn are presented in Figure 6. The mean concentrations of Cd in the cores range from 0.06 mg/kg (SA PIT1) to 0.14 mg/kg (Albania). These con- centrations correspond to the median of 0.11 mg/kg given by DE LAZZARI et al. (2004), although their data shows large variability across the northern and central Adriatic with a concentration range from 0.04 to 0.97 mg/kg. The highest concentration of Cd determined in the Albania core was in the deeper part of the core (22–24 cm). Total mercury concentrations in the sediment cores range from 13 to 106 μg/kg, with the mean concentration between 26 and 34 μg/kg. The Hg depth profiles show a clear upward increase in the analyzed cores (Figure 6), with the exception of the Palagruža core where Hg slightly decreases in the top section. The lower sections of the cores (SA PIT, Palagruža, Jabuka) have low Hg concentrations in a narrow range be- tween 13 and 23 μg/kg. The Albania core has higher contents of Hg in the lower sections of the core ranging from 33–50 μg/kg. The Jabuka and Palagruža cores have generally slightly higher contents of Hg in comparison to the South Adriatic Pit cores (Figure 6). Pb concentrations show a general increase in the upper core samples (Figure 6), there is a wide peak in the SA core between 30–32 cm depth. This interval also contains elevated concentrations of other trace elements (Ba, Rb) and magnetic Figure 3: Distribution and abundance of calcite and quartz in the studied cores (SA PIT 1 and 2 are merged into SA PIT). Figure 4: A representative (SA PIT 1 6-8 cm) oriented sample of the clay mineral fraction, with basal reflections of clay minerals after treatments (AD-air dried, EG-ethylene glycolated, heated to 400°C and heated to 550°C). Symbols: Sm-smectite, Chl-chlorite, Ill-illite, Kln-kaolinite. Ilijanić et al.: Metal deposition in deep sediments from the Central and South Adriatic Sea Geologia Croatica 191 susceptibility that could possibly be related to tephra. There is a slight decrease of Pb in the tops of the Palagruža and Albania cores. In all the cores the Pb concentrations exhibit slight variation from 14.9 to 24.5 mg/kg, with mean values ranging from 18.1 to 20.2 mg/kg (Supplementary Table 1). Down-core variations in Cr and Ni are not pronounced and the difference between the cores (Supplementary Table 1, Figure 6) lies in the total contents along the EAC transect from the Albania core with the high concentrations of Cr (mean 263 mg/kg), and the lowest in the Palagruža core (120 mg/kg). The mean concentrations of Cr in the SA PIT cores (152 and 146 mg/kg) and the Jabuka core (131 mg/kg) falls between these two. The concentrations of Ni show exactly the same behaviour (r=0.99, Supplementary Table 2, Figure 6). The lower Ni and Cr contents in the Palagruža core are a consequence of carbonate dilution, since the Palagruža cores contain more calcite than the other cores (Table 2., Figure 5). The down core distributions of Mn, As, Mo and Cu are quite complex in the analyzed cores especially in the South Adriatic Pit. The concentrations of Mn in the analyzed cores range from 700 mg/kg (Palagruža) to 5201 mg/kg (SA PIT). The down-core distribution of Mn in the South Adriatic pit substantially differs from that of the other core exhibiting the highest concentration (5201 mg/kg) in the 18–20 cm in- terval and a peak (3942 mg/kg) in the 32–34 cm interval (Fig ure 6). The As concentrations in the analyzed cores were gen- erally higher than those reported by DOLENEC et al. (1998) for the southern Adriatic. The Palagruža core samples (mean 7.4 mg/kg) and the Jabuka core samples (mean 13.2 mg/kg) were within the range of the As proportions in sediments in the Gulf of Manfredonia (SPAGNOLI et al., 2008). If we exclude the anomalous concentrations (64 mg/kg) measured in the bottom part of the SA PIT core (Figure 6), the highest mean concentrations were in the Albania core (22 mg/kg). High As concentrations occur in the samples from 30–38 cm depth (Figure 6). Cu concentrations in the cores show similar ranges in the SAP PIT and Albania cores with a narrow average 38.9 mg/kg and 41.1 mg/kg (Supplementary Table 1). The mean Cu contents of both the Palagruža and Jabuka are lower (21.6 and 29.4 mg/kg, respectively). The positive correlation with Al (r=0.83) is indicative, (as in case of the analyzed transi- tion metals) that concentration is related to detrital alumino- silicates. Similar values for the Adriatic surface sediments were discovered by DOLENEC et al. (1998). The concentration of Zn in the analyzed cores ranges on average from 68 mg/kg in the Palagruža core to 105 mg/kg Figure 5: Relationships between the major elements (Ca, Al, Ti) and selected mineral phases (calcite and muscovite/illite). Geologia Croatica 67/3Geologia Croatica 192 Figure 6: Down-core variability of Pb, Hg, Zn, Cd, Cr, Ni, Cu, Mn, and Mo in the studied sediment cores. Ilijanić et al.: Metal deposition in deep sediments from the Central and South Adriatic Sea Geologia Croatica 193 Figure 6: Continued. Supplementary Table 1: Metal concentrations ranges, mean, and standard deviation (std) in the analyzed cores.   Core JABUKA PALAGRUŽA SA PIT 1 SA PIT 2 ALBANIA Cd (mg/kg) mean 0.094 0.094 0.087 0.084 0.122 std 0.011 0.011 0.015 0.009 0.016 min. 0.080 0.080 0.060 0.070 0.100 max. 0.110 0.110 0.120 0.090 0.140 Hg (ug/kg) mean 34 31 27 26 55 std 12 5 11 15 29 min. 18 23 13 13 33 max. 51 36 60 51 106 Mo (mg/kg) mean 1.36 0.70 1.24 1.66 0.84 std 0.92 0.30 0.51 0.51 0.27 min. 0.50 0.40 0.50 1.10 0.60 max. 2.40 1.20 2.60 2.50 1.30 Cu (mg/kg) mean 29.4 21.6 38.9 39.2 41.1 std 2.2 2.8 3.9 3.3 2.3 min. 26.5 19.3 30.6 34.9 37.8 max. 32.5 26.4 44.5 43.9 44.2 Pb (mg/kg) mean 18.1 17.5 18.8 20.0 20.2 std 2.0 2.8 2.1 2.7 1.7 min. 16.2 14.9 16.6 16.7 18.3 max. 20.7 20.7 24.5 23.9 22.2 Zn (mg/kg) mean 85 68 85 83 105 std 5 6 5 3 4 min. 79 61 78 79 102 max. 91 75 102 87 112 Cr (mg/kg) mean 131 120 152 146 263 std 6 6 7 8 9 min. 123 112 138 138 251 max. 138 128 164 157 276 Geologia Croatica 67/3Geologia Croatica 194 in the Albania core. The high positive correlation with Cr (r=0.83, Supplamentary Table 2) is indicative of a similar source related to detrital aluminosilicates. 5. DISCUSSION 5.1. Sedimentation rates The natural and artificial radionuclide activity concentra- tions in the sediments from the central and south Adriatic Sea are comparable with other reported values from the Me- diterranean (OTHMAN et al. 2000) and Adriatic Sea (FRI- GNANI et al., 2004, 2005; ŠTROK et al., 2010). In addition, their values in seawater and sediments throughout the East- ern Adriatic coast indicate that the determined values do not pose a significant risk for most marine biota (PETRINEC et al., 2013). The Adriatic Sea sedimentation rates have been mainly studied in the North Adriatic in which sedimentation is highly influenced by the Po River (PALINKAS & NIT- TROUER, 2007). Therefore, sedimentation in that area is quite high, and the sedimentation rate was estimated to be between 5–18 mm/y (FRIGNANI et al, 2004.). Sedimentation rates estimated for the Mid and South Adriatic using 137Cs as a radiotracer are consistent with sed- imentation rates estimated for the rest of the Mediterranean sea, i.e., 1.1–8.7 mm/y (OTHMAN et al., 2000) using other radiotracer methods. PALINKAS & NITTROUER (2007) showed that sediment accumulation on the Po shelf is largely controlled by the flood sedimentation near the mouth of the main tributaries, with the highest sedimentation rates of 10- 40 mm/y, while in the southern area of the dispersal system, accumulation is lower (< 10 mm/y), reflecting sedimentation during non-flood periods. The available data for the Adriatic Sea (FRIGNANI et al., 2005) shows that about 45% of val- ues of the accumulation rates are lower than 2 mm/y. 5.2. Mineral composition The mineral composition of the analysed cores shows good agreement with CALANCHI et al. (1996) who described the main mineral phases in the Adriatic sea sediments as quartz, calcite, dolomite, feldspars, micas, clay minerals (musco- vite-illite, smectite and chlorite) with occasional occurrences of amphiboles and serpentine. Furthermore, several authors (PIGORINI, 1968; NELSON, 1972; TOMADIN, 2000) an-   Core JABUKA PALAGRUŽA SA PIT 1 SA PIT 2 ALBANIA Ni (mg/kg) mean 149 122 179 175 315 std 7 8 8 7 10 min. 139 109 162 168 299 max. 156 130 192 182 323 As (mg/kg) mean 13.2 7.4 24.7 18.0 22.0 std 5.2 5.1 14.0 2.1 4.4 min. 5.0 3.0 15.0 15.0 18.0 max. 18.0 16.0 64.0 20.0 29.0 Mn (mg/kg) mean 1990 892 1826 2059 820 std 1125 397 1181 900 131 min. 1120 700 882 1362 732 max. 3457 1602 5201 3156 1047 Ca (%) mean 11.06 15.02 10.77 10.66 5.02 std 0.95 1.36 0.61 0.48 0.14 min. 9.95 13.69 9.27 10.05 4.83 max. 12.29 16.81 11.63 11.11 5.19 Ti (%) mean 0.23 21 0.27 0.27 0.36 std 0.01 0.01 0.01 0.01 0.01 min. 0.22 0.21 0.25 0.26 0.34 max. 0.25 0.22 0.29 0.27 0.37 Al (%) mean 5.73 4.94 6.12 6.06 6.56 std 0.16 0.19 0.20 0.20 0.34 min. 5.56 4.69 5.71 5.87 6.11 max. 5.98 5.14 6.59 6.36 6.92 Supplementary Table 1: Continued. Ilijanić et al.: Metal deposition in deep sediments from the Central and South Adriatic Sea Geologia Croatica 195 alyzed the clay mineral distribution in superficial sediments of the central and southern Adriatic Sea. They reported that in the Mid Adriatic Depression, clay minerals of Apenninic origin dominate, whereas in the southern Adriatic Sea clay minerals show a depth-controlled distribution, suggesting strong hydrodynamic selection which is indicated by rela- tive quartz dilution in offshore cores in comparison with the near shore Albania core (Figure 3). The distributions of il- lite, chlorite, and quartz seem not to be depth related but are a consequence of the carbonate content of the sediments. Apennine sediments (pelitic sediments rich in smectite) are dispersed transversally to the central basin and overlie the Padane, longitudinal, southward dispersion of fine-grained sediments (illite- and chlorite-rich sediments). Smectite (which is the dominant clay mineral in all the analysed sed- iment cores), shows the afore mentioned non depth-control- led distribution with higher mean contents in the Jabuka and South Adriatic Pit (25%) and lower (20%) mean contents in the shallower Palagruža and Albania cores. This could be in accordance with the findings of TOMADIN (2000) and due to the southern Adriatic basin turbidity currents which flow transversely to the basin carrying clay sediments from the Apulian shelf and the Albanian–Montenegrin shelf into the bathyal basin. TOMADIN (2000) used the clay mineral com- position and distribution in the central and southern Adria tic Sea as tracers of sediment provenance, indicating longitudi- nal dispersion connected with the general Adriatic cyclonic circulation. He recognized fluxes directed south-westward along the Italian offshore – the Apennine flux near the coast (along the WAC) and the Padane flux in the open sea; and a northenward flux, along the eastern coasts of the central and southern Adriatic basin, the Albanian flux (along the EAC). Based on the contents of illite, chlorite, quartz and calcite along the EAC (Table 2, Fig. 3) a trend is observed indicat- ing an enrichment of calcite and depletion of the other three minerals along the EAC. 5.3. Geochemical considerations 5.3.1. Major elements The down-core contents of the major elements Ca, Al and Ti display narrow ranges in the individual cores. These ele- ments and their relationship to trace elements are important in evaluating possible pollution. Al, Ti as well as Zr, Sr Li, Rb are considered as conservative during chemical weather- ing (SHOTYK et al., 2001; VAN DER WEIJDEN, 2002). Most methods used assume aluminum as present only in clays and detrital aluminosilicates (VAN DER WEIJDEN, 2002) and the metals are attributed to the clay fraction. Ti- tanium is commonly hosted in high specific gravity minerals that are not easily transported aerially (Reimann & DE Car- itat, 2005) and indicates detrital, coarser fractions, deposited in higher energy environments (SPAGNOLI et al., 2008). In Supplementary Table 2: Spearman correlation matrix (qtz-quartz, cal-calcite, k-fs-potassium feldspar, ms/i-muscovite/illite, sm-smectite, chl-chlorite). Cd Hg Mo Cu Pb Zn Ni Mn As Ca Cr Ti Al qtz cal k-fs ms/i sm chl Cd 1.00 Hg 0.42 1.00 Mo 0.02 0.06 1.00 Cu 0.21 0.15 0.37 1.00 Pb 0.43 0.51 0.41 0.53 1.00 Zn 0.51 0.51 0.03 0.65 0.49 1.00 Ni 0.56 0.41 –0.13 0.56 0.25 0.83 1.00 Mn –0.12 –0.23 0.63 0.08 –0.06 –0.13 –0.17 1.00 As 0.17 0.38 0.33 0.74 0.40 0.62 0.59 0.22 1.00 Ca –0.56 –0.50 –0.08 –0.67 –0.42 –0.93 –0.91 0.06 –0.67 1.00 Cr 0.58 0.47 –0.21 0.48 0.23 0.81 0.99 –0.24 0.54 –0.88 1.00 Ti 0.48 0.32 –0.10 0.62 0.22 0.81 0.98 –0.10 0.61 –0.89 0.95 1.00 Al 0.24 0.11 0.19 0.83 0.27 0.76 0.74 0.06 0.70 –0.79 0.67 0.81 1.00 qtz 0.46 0.38 –0.23 0.43 0.24 0.66 0.69 –0.41 0.34 –0.61 0.68 0.62 0.54 1.00 cal –0.41 –0.35 –0.13 –0.59 –0.36 –0.75 –0.78 –0.07 –0.56 0.84 –0.76 –0.81 –0.69 –0.22 1.00 k-fs 0.43 0.13 –0.14 0.07 0.07 0.16 0.36 –0.24 0.00 –0.24 0.38 0.34 0.00 0.10 –0.27 1.00 ms/i 0.65 0.44 –0.17 0.35 0.33 0.57 0.56 –0.42 0.10 –0.56 0.58 0.51 0.40 0.69 –0.35 0.19 1.00 sm –0.51 –0.32 0.34 –0.01 –0.17 –0.22 –0.33 0.61 0.11 0.21 –0.39 –0.24 0.01 –0.65 –0.11 –0.27 –0.63 1.00 chl 0.14 0.20 0.02 0.19 0.08 0.38 0.48 0.07 0.29 –0.48 0.50 0.51 0.30 –0.18 –0.80 0.29 –0.04 0.25 1.00 Geologia Croatica 67/3Geologia Croatica 196 the analyzed cores, Ti has very high correlation coefficients with Ni (r=0.98) and Cr (r=0.95) and correlates well with muscovite/illite (r=0.65) and quartz (r=0.62), indicating a presence in the silty fractions (Supplementary Table 2) (SPAGNOLI et al., 2008). The relationships between Ti, Al, quartz and muscovite/illite are illustrated in Figure 5. Ca is mainly derived from marine biogenic carbonate and Al from detrital aluminosilicates (DOLENEC et al., 1998 and GOU- DEAU et al., 2013). This is confirmed by a high correlation (r=0.8) between Ca and calcite content indicating that most of Ca is derived from calcite (Supplementary Table 2.). 5.3.2. Trace elements The analyzed trace elements based on their distributions can be divided into 4 groups; a) the “pollution”-related elements Pb, Hg, and Cd (REIMANN & DE CARITAT, 2005) show- ing a distinct concentration increase in the 0–2 cm core in- tervals (Figure 6); b) Cr and Ni which are the most com- monly used trace elements for sediment provenance studies in the Adriatic basin (AMOROSI et al., 2002; DINELLI et al., 2007; LUCCHINI et al., 2003; SPAGNOLI et al., 2010; AMOROSI, 2012; GOUDEAU et al., 2013); c) Zn, Cu, As that display a complex distribution probably related to the clay and detrital aluminosilicate fractions with relatively high correlation (r>0.7) with Al (Supplementary Table 2). The distribution of As and Cu could also partially be con- trolled by redox processes (NAMEROFF et al., 2002), but due to the lack of correlation with Mn and Mo they are treated as a separate group: d) redox-sensitive trace metals Mn and Mo which were found to be significant features of the bottom sediments of MAD (DOLENEC, 2003) but also in numerous shallow marine sediments (ANSCHUTZ et al., 2005; BECK et al., 2008). Lead and mercury – DOLENEC et al. (1998) suggested Pb concentration ranges for the central Adriatic sediments from 5–18 mg/kg and 7–14 mg/kg. SPAGNOLI et al. (2010) in their overview of trace element data cited 34 mg/kg as a background value for the Mid Adriatic Depression and 9.76 mg/kg for the Albanian coast. ROMANO et al., 2013 deter- mined that sediments along the WAC had mean Pb concen- trations ranging from 26–48 mg/kg. It is considered that Pb emissions are a result of many human activities including coal combustion, use of leaded gasoline, metal smelting, and mining. EICHLER et al. (2012) gave historical Pb emis- sion estimates for the former U.S.S.R. Countries which con- siderably increased after 1935 with maximum values between 1970 and 1975. After this peak period Pb concen- trations subsequently decreased and returned in the 1990s to the level of 1940−1950 (EICHLER et al., 2012). In the analyzed cores from the Adriatic the slight decrease of lead in the surface (0–2 cm) intervals in the Albania core could be interpreted as a consequence of this Pb emission de- crease. ANNIBALDI et al. (2009) determined that Pb con- centrations in Adriatic Sea water have shown a substantial decrease since 2001, as a consequence of decreased Pb emis- sions to the atmosphere in Italy. The absence of this dimi- nution in the Mid Adriatic Depression (Jabuka) and the Southern Adriatic Pit (SA PIT1) surface sediments could be explained by substantially lower sedimentation rates in the depressions, and that analysis of 2 cm intervals did not give adequate temporal resolution. The Hg depth profiles show a clear upward increase in the analyzed cores and that the background Hg concentration values for deep Adriatic sediments are less than 25 μg/kg. MASON et al. (2012) stressed that anthropogenic activities have enriched Hg in the biosphere by at least a factor of three, and that it is supplied to the offshore regions in the form of wet and dry atmospheric deposition. The background values for the northern Adriatic (Gulf of Trieste and Po delta) given by COVELLI et al. (2006) and RAJAR et al. (2006) are ap- proximately 130 μg/kg. DOLENEC et al. (1998) concluded that most of the Hg supplied to the central Adriatic was de- rived from the northern Adriatic with minor influences from the Croatian coast, where Kaštela bay sediments near Split could be a possible localized source (LOVRENČIĆ MI KE- LIĆ et al., 2013). SPAGNOLI et al. (2010) cite a range of 67–224 μg/kg, for the Mid Adriatic Depression and 17 μg/kg, as the background for the coastal sediments of Albania. Chromium and nickel – Most authors for provenance studies in the Adriatic basin use Cr and Ni to define sources of sed- iment to the Adriatic along the Western Adriatic Current and its mud belt (AMO ROSI et al., 2002; LUCCHINI et al., 2003; DINELLI et al., 2007; WELTJE & BROMMER, 2011; AMOROSI, 2012; GOUDEAU et al., 2013) influenced by sediment discharge from the Po river and the Apennine riv- ers. The Ofanto River contributes most to the coastal sedi- ments south of the Gargano promontory along the south east- ern coast of Italy (GOUDEAU et al., 2013). GOUDEAU et al. (2013) concluded that a negligible contribution of min- eral matter from the eastern Adriatic coast is possible. Using a chemostratigraphical approach in determining the spatial and temporal supply in long Adriatic cores from the Adriatic Shelf, the MAD and the South Adriatic Pit, LUCCHINI et al. (2003) showed that Cr and Ni were supplied by the Po River. The Holocene sediments of MAD have mean concen- trations of 191 mg/kg Cr and 195 mg/kg Ni while the Last Glacial Maximum (LGM) sediments have lower mean con- centrations 117 mg/kg Cr and 79 mg/kg Ni (LUCCHINI et al., 2003). The analyzed cores mainly lie along the western side of the Adriatic and the influence by the EAC is probably insignificant (GOUDEAU et al., 2013). The mean concen- trations of Cr and Ni for the southern and central Adriatic given by DOLENEC et al. (1998) are lower than those in the analyzed cores but plot within the concentration ranges. The concentrations are dependent on the carbonate dilution i.e. the content of calcite in the sediment. They have high nega- tive correlation coefficients with calcite content (Supplemen- tary table 2) and are positively correlated with the muscovite/ illite phase. High Cr and Ni proportions in the Albania core are most probably a consequence of weathering and erosion of catch- ments containing ophiolitic mélanges associated with ophi- olitic sequences which are part of the Mirdita–Subpelagonian zone. High concentrations of transition metals (such as Ni, Co, Cr and V) are one of their primary geochemical features Ilijanić et al.: Metal deposition in deep sediments from the Central and South Adriatic Sea Geologia Croatica 197 (SACCANI & PHOTIADES, 2005). The link with detrital aluminosilicates is displayed by correlation with both musco- vite/illite, chlorite mineral phases (Supplementary Table 2). A contribution of Ni and Cr is also possible from the Dalmatian region in the form of soil dust transported by the strong Bora wind especially to the eastern side of the MAD, or colloid suspensions from the eastern Adriatic rivers. Based on the geochemical soil data (MIKO et al., 2001; HALAMIĆ et al., 2012), soils developed on karst in Dalmatia contain high concentrations of Cr (mean 119 mg/kg) and Ni (mean 79 mg/kg) and due to the lack of vegetation cover are sus- ceptible to wind and water erosion. Zinc, Copper and Arsenic – Although Zn, Cu and As are of- ten considered as indicators of possible anthropogenic influ- ences, the high positive correlation with Cr (Supplementary Table 2) is indicative of a similar source related to detrital aluminosilicates. The concentrations fall within the ranges given by DOLENEC et al. (1998) and DE LAZZARI et al. (2004) for the central and southern Adriatic and SPAGNOLI et al. (2008) for the Gulf of Manfredonia. Also the concen- trations are dependent on the carbonate dilution i.e. the con- tent of calcite in the sediment as in the cases of Ni and Cr in the Palagruža core. Concentrations of As in the analyzed cores are generally higher than those reported by DOLENEC et al. (1998) for the southern Adriatic, and slightly lower than those for the sediments in the Gulf of Manfredonia given by SPAGNOLI et al. (2008). Manganese and molybdenum – The concentrations of Mn are comparable with those of DOLENEC et al. (1998) from the Jabuka pit and the southern Adriatic. The measured con- centration of 3457 mg/kg is similar to the 3760 mg/kg of DOLENEC et al. (1998). Mn in the southern Adriatic ac- cording to DOLENEC et al. (1998) could have been derived from the detrital material from the Albanian hinterland. DE LAZZARI et al. (2004) also determined that the Mn propor- tion was much higher in the MAD than other areas of the central and northern Adriatic. The Albanian core has the low- est concentration of Mn (732–1047 mg/kg). With the excep- tion of the SA PIT cores the other three cores show relative surface increases in Mn concentrations (Figure 6). DO- LENEC (2003) described the manganate crusts at the sedi- ment-water interface in the Jabuka pit and their genesis as a result of processes linked with oxic diagenesis and the ob- servation that sediments below 5 cm contain considerably less Mn. ANSCHUTZ et al. (2005) and CHAILLOU et al. (2008), showed that in the first few centimetres of the sedi- ment, Mn undergoes several redox cycles with numerous possible reactions generally resulting in an aerobic surface layer enriched in Mn (III, IV) phases, as in the case of the Albania, Palgruža and Jabuka cores. At deeper levels, these oxides are reduced to soluble Mn (II) (CHAILLOU et al., 2008). The vertical distributions of Mo follow a similar pat- tern with Mn in the Albania, Palagruža and Jabuka cores which are a consequence of the high affinity of the molyb- date (MoO4 2−) ion, (the soluble form of Mo in oxic water), with Mn-oxides in marine sediments (CHAILLOU et al., 2008). These features are also revealed from the association of Mo and Mn in other sediments and ferromanganese nod- ules. The sediment geochemistry of Mo may indicate short- term variations in pore water chemistry, or the influence of Mn diagenesis (MORFORD & EMERSON, 1999). The de- coupled distributions of Mn and Mo (Figure 6) indicate that in the SA PIT an anoxic horizon is present in the 2–10 cm interval. This could have led to a partial decrease in Mn con- centrations and authigenic enrichment of Mo as a conse- quence of the reduction of Mo (VI) to Mo (IV) due to sulfate reduction (CHAILLOU et al., 2008). In marine sediments Mo authigenesis is controlled by authigenic sulphide phases (CHAILLOU et al., 2008). The formation of Mn (oxyhy- droxides-oxides?) enriched layers in the analyzed cores is to a degree favoured by the presence of smectite (r =0.61) where their formation is probably as coatings on the clay minerals which are considered to be uniformly negatively charged over their (001) crystal faces (FÖRSTNER & WITT- MAN, 1983). 5.4. Normalization and enrichment factors Trace metals may be introduced in the surface layers of ma- rine sediments as a result of both natural processes (e.g. che- mical weathering and erosion) and human activities within the catchment. The trace metals are mainly contained within the inorganic material which consists of silicate minerals such as quartz, feldspar, but more relevant micas and clay minerals and smaller amounts of metal oxides and sulfide phases. Differing contents of trace element and heavy met- als in sediments do not necessarily indicate varying degrees of pollution, but could reflect dissimilarities in grain size and mineralogical composition or post depositional remobiliza- tion. Normalization with respect to characteristic elements used as proxies for grain size or differing sediment sources is often used to distinguish anthropogenic pollution effects from natural variability. A variety of approaches for reduc- ing natural variability have been used to improve the statis- tical power in data inter-comparison (WINDOM et al., 1989; VAN DER WEIJDEN, 2002; REIMANN & DE CARITAT, 2005). There is no consensus on the appropriate sediment component to be used for normalization i.e. to factor out the variability in natural trace element concentrations. In most cases, the source of the natural material making up the sed- iment has been assumed to be constant so the emphasis has been placed on accounting for the “grain size effect” (WIN- DOM et al., 1989). Metals are not homogeneously distrib- uted over the various grain size fractions, and large differ- ences in total concentrations are observed in sediment samples from a single locality. Within the grain size spec- trum, the finer-grained fraction consisting mainly of clay minerals, show relatively high metal contents. In the silt and fine sand fractions the metal concentrations generally de- crease as those fractions are dominated by quartz and feld- spars with low metal contents, although the occurrence of heavy minerals may cause an increase in selected elements (VITAL & STATTEGGER, 2000; GARCIA et al., 2004; DINELLI et al., 2007). Geologia Croatica 67/3Geologia Croatica 198 The basic geochemical approach is to normalize geo- chemical data by means of a conservative component the levels of which are unaffected by contaminant inputs, for example, grain size, Al, Fe, Sc, Ni, TOC and Li. Conse- quently, changes/dilutions by the CaCO3, silica, or organic matter content, especially in the upper layers, can be cor- rected (FÖRSTNER & WITTMANN, 1981) and be used for comparison with standard reference materials such average shale, upper crust (SHOTYK et al., 2001; VAN DER WEI- JDEN, 2002). This approach has several limitations, and was not used in this study. In particular, concentrations for crus- tal abundances are not appropriate because they do not rep- resent regional background levels because element ratios dif- fer substantially from rock type to rock type and the ratios in actual rocks do not reflect calculated or modeled crustal ratios (REIMANN & DE CARITAT, 2005). Typical normalization approaches identify one or several reference elements as a most frequently used conservative element. Aluminum is the most commonly used conservative normalization for trace element sediment data (VAN DER WEIJDEN, 2002, DI LEONARDO et al., 2006, SPAGNOLI et al., 2008), alternatively this element is replaced by Sc (SHOTYK et al., 2001). The concentration of normalizing metal is used to establish the relationship between natural trace metal concentrations in sediments from different areas. The normalization procedure with Al as a conservative ele- ment involving transitional elements was also widely used for provenance studies in the Adriatic basin. Here, Cr/Al and Ni/Al were useful geochemical tracers used to define sources of sediment to the Adriatic along the Western Adriatic Cur- rent and its mud belt (AMOROSI et al., 2002; LUCCHINI, et al.; 2003, DINELLI et al.; 2007; SPAG NOLI et al., 2008; AMOROSI, 2012; GOUDEAU et al., 2013). The normaliza- tion of element contents to a reference element was applied to compensate for grain size and mineralogical effects. Al and Ti are generally used to estimate the percentage of terrestrial materials in marine sediments, and to subtract the contribu- tion of detrital materials when calculating the authigenic com- ponents of marine sediments. For both pelagic and terrige- nous materials-dominat ed se diments, the Ti/Al ratio has also been found to reflect the average grain size of silicate miner- als in marine sediments (SHOTYK et al., 2001). Elevated heavy metal concentrations in the surface lay- ers of marine sediments are also sometimes attributed to an- thropogenic heavy metal input to the environment (VAN DER WEIJDEN, 2002). In order to verify possible anthro- pogenic input regarding heavy metals, enrichment factors of selected elements were calculated. Also the EFs were calcu- lated for Ni and Cr to in order to identify possible temporal changes in the supply of these elements. Figure 7: Down-core variability of Metal/Al-normalized values in the studied sediment cores. Ilijanić et al.: Metal deposition in deep sediments from the Central and South Adriatic Sea Geologia Croatica 199 The estimation of heavy metal enrichment (VAN DER WEIJDEN, 2002) was made by a calculation procedure which uses concentrations of heavy metals and conservative elements from deeper layers (RIDGWAY & SHIMMIELD, 2002) as background reference for the uppermost sediments. The EF is a concentration ratio of a given element (Cnsample) to the conservative element (Al in this study) in the sample (Ccons.sample) with respect to the same ratio in the reference Table 3: Concentration values and enrichment factors (EF) for Pb, Hg, Cd, Zn, Cr and Ni in the top sections (0-2 cm) of the cores. Core Sediment interval (cm) Depth (m) Pb   Hg   Cd   Cu   (mg/kg) EF (mg/kg) EF (mg/kg) EF (mg/kg) EF JABUKA 0-2 230.8 20.7 1.3 51 2.9 0.11 1.3 31.78 1.1 PALAGRUŽA 0-2 169.8 20.7 1.3 32 1.4 0.11 1.4 25.48 1.4 SA PIT 1 0-2 1041.4 24.5 1.4 60 2.3 0.12 1.6 44.59 1.3 SA PIT 2 0-2 1030 23.9 1.4 51 2.6 0.09 1.4 42.51 1.4 ALBANIA 0-2 59 20.4 1.3 106 3.6 0.1 0.9 31.82 1.0     Core Sediment interval (cm) Depth (m) Ni   Cr   Zn     (mg/kg) EF (mg/kg) EF (mg/kg) EF   JABUKA 0-2 230.8 141.3 0.9 77 0.9 85 1.1   PALAGRUŽA 0-2 169.8 114.5 1.0 67 0.9 75 1.1   SA PIT 1 0-2 1041.4 167.2 0.9 91 1.0 102 1.2   SA PIT 2 0-2 1030 164.2 1.0 94 0.9 85 1.1   ALBANIA 0-2 59 268.2 1.1 152 1.1 103 1.1     Supplementary Table 3: Mean values of metal/Al ratios in surface sediments and cores along the Eastern Adriatic Current (EAC) transect going from AL- BANIA to the Gulf of Trieste (NAS Tri), and along the Western Adriatic Current (WAC) transect along the coast of Italy to the southern Adriatic. Expl. AL- BANIA, SAPIT1, SAPIT2, PALAGRUŽA, JABUKA cores from this study. Data from DOLENEC et al., (1998) was classified as follows SAS ALB south Adri- atic sediments along the Albanian coast, SAS Cro south Adriatic sediments along the Croatian coast, CASJABUKA sediments from the Meso-Adriatic Depression, CASCro central Adriatic sediments along the Croatian coast, NASCroa northern Adriatic sediments along the Croatian coast, NASTri sedi- ments from the Gulf of Trieste, NASPo northern Adriatic sediments under the influence of the Po River, NASap northern Adriatic sediments along the eastern coast of Italy, SASap southern Adriatic sediments along the coast of Italy. The data shown for Gargano peninsula and the Gulf of Manfredonia is from GOUDEAU et al. (2013). Ca/Al Al/Ti Cr/Al Ni/Al Hg/Al Pb/Al Zn/Al Cu/Al As/Al ALBANIA* EAC 0.77 18.2 40.1 48.0 155 3.09 16.1 6.27 3.38 SAS ALBa 1.69 20.9 20.6 1.0 337 1.88 14.5 7.51 1.58 SA PIT2* 1.76 23.0 24.8 29.3 100 3.07 13.9 6.36 4.05 SA PIT1* 1.76 22.7 24.1 28.8 97 3.31 13.7 6.48 2.97 SAS Croa 2.17 21.5 33.9 36.1 448 2.74 16.0 5.95 3.62 PALAGRUŽA* 3.05 23.0 24.4 24.8 143 3.54 13.8 4.38 1.49 JABUKA* 1.93 24.4 22.8 25.9 149 3.16 14.8 5.13 2.31 CAS JABUKAa 2.43 22.8 26.0 26.3 677 2.21 16.3 5.52 0.51 CAS Croa 4.57 20.8 35.6 33.4 525 3.49 17.3 5.69 4.72 NAS Croa 4.42 13.6 19.2 9.1 1737 4.26 14.0 3.02 1.03 NAS Tria WAC 2.51 19.3 18.1 13.1 1729 5.03 20.8 5.59 0.99 NAS Poa 2.90 20.7 17.6 9.7 1451 5.15 19.1 3.51 0.61 NAS Apa 2.26 23.0 13.8 8.0 2959 2.30 15.5 4.71 0.52 CAS Apa 2.83 22.5 19.9 18.7 775 2.49 16.0 5.61 3.28 Garganob 1.65 nd 19.1 13.9 nd nd 9.7 4.20 nd Manfredoniab 1.87 nd 18.9 12.4 nd nd 10.8 4.30 nd SAS Apa 2.49 21.2 12.4 10.9 612 2.25 13.8 6.60 0.60 *This study, a DOLENEC et al., 1998; b GOUDEAU et al., (2013); nd-no data. Geologia Croatica 67/3Geologia Croatica 200 material Cn ref./Ccons. ref. (FÖRSTNER & WITTMANN, 1981; LI, 1981): EF = (Cnsample/Ccons.sample ) / (Cn ref./Ccons. ref.). Five degrees of contamination are commonly defined (SUTHERLAND, 2000): EF < 2: deficiency to low enrich- ment; EF 2–5: moderate enrichment; EF 5–20: significant enrichment; EF 20–40: very high enrichment; EF > 40: ex- Figure 8: Box plots of metal/Al ratios in surface sediments and cores along the Eastern Adriatic Current (EAC) transect going from ALBANIA to the Gulf of Trieste (NAS Tri), and along the Western Adriatic Current (WAC) transect along the coast of Italy to the southern Adriatic. Expl. ALBANIA, SAPIT (1&2), PALAGRUŽA, JABUKA cores from this study. Data from DOLENEC et al., (1998) was classified as follows SAS ALB south Adriatic sediments along the Al- banian coast, SAS Cro south Adriatic sediments along the Croatian coast, CASJABUKA sediments from the Meso-Adriatic Depression, CASCro central Adriatic sediments along the Croatian coast, NAS Cro northern Adriatic sediments along the Croatian coast, NAS Tri sediments from the Gulf of Trieste, NAS Po northern Adriatic sediments under the influence of the Po River, NASap northern Adriatic sediments along the eastern coast of Italy, SAS Ap southern Adriatic sediments along the coast of Italy. The data shown for Gargano peninsula and the Gulf of Manfredonia are from GOUDEAU et al., (2013). Median 25%-75% Non-Outlier Min-Max Ilijanić et al.: Metal deposition in deep sediments from the Central and South Adriatic Sea Geologia Croatica 201 tremely high enrichment. The down-core variations of Pb/ Al, Hg/Al, Zn/Al, Cr/Al and Ni/Al (Figure 7) in general show some smoothing of the profiles, in comparison with elemen- tal distributions. Hg, and Pb show distinct higher values in the top sediment intervals. Generally the Cr/Al and Ni/Al ratios decrease from the Albania core (40.1 and 48.0 respec- tively) along the EAC transect to the Jabuka core (22.8 and 25.9 respectively; Supplementary Table 3, Figures 7 and 8). Since the metal/Al ratios can be applied to determine metal excesses or to account for natural changes, they were used to give an insight to the changes in trace metal proportions along the transect of the Eastern Adriatic Current (EAC) from Albania, along the Croatian coastal islands, through the Northern Adriatic and then along the Western Adriatic Cur- rent (WAC) to the south. An overview of the provenance of surface sediments along the WAC on the southeastern coast of Italy was given by GOUDEAU et al. (2013). They used a multi-proxy approach which included Cr/Al and Ni/Al ra- tios. These data were used together with the ratios calculated from the data of DOLENEC et al. (1998). Ratios of Cr and Ni (the trace metals used in provenance studies), (AMOROSI et al., 2002; LUCCHINI, et al., 2003; DINELLI et al., 2007; SPAGNOLI et al., 2008; AMOROSI, 2012; GOUDEAU et al., 2013), are significantly different (higher) in the southern and central Adriatic along the EAC (Figures 8 and 9), compared to those of the surface sediments of the Northern Adriatic and along the WAC (lower). GAU- DEAU et al. (2013) give a Cr/Al ratio of 27.3 for the Po River and an approximate value of 15.5 for sediments influ- enced by the Apennine rivers. The Gargano peninsula sedi- ments and the Gulf of Manfredonia have mean higher Cr/Al ratios of 19.1 and 18.1, respectively. The values of these ra- tios are similar to those given by LUCCHINI et al. (2003) for the Adriatic shelf cores of both Holocene and Pliocene age. The Cr/Al ratios for Holocene sediments from the MAD and SA cores given by LUCCHINI et al. (2003) have mean values higher than 23 that correspond to the Cr/Al ratios in the SA PIT cores (24.8), Palagruža core (24.4) and the Jabuka core (26.0). The Pleistocene sediments from the cores in the Mid Adriatic Depression have Cr/Al ratios lower than 20 and are comparable to the Adriatic shelf Cr/Al ratios (LUC- CHINI et al., 2003). The high Cr/Al (40.1) ratio from Alba- nia, which is a signature of weathering and erosion of catch- ments Mirdita ophiolitic sequences, is carried as the “Alba nian flux” (TOMADIN, 2000) along the EAC. The plotted Cr/Al vs Ni/Al ratios (Figure 9) for the analysed cores and the data from DOLENEC et al. (1998) and GOUDEAU et al. (2013) showed that a linear dilution path between Al- bania and the Apennine (WAC) signature is distorted. This indicates the possible influence of another source that is de- creasing the ratios, possibly from the Dinaric/Dalmatian source of minerals. The distortion could also be attributed due to preferential changes in grain size towards finer frac- tions which were not compensated by the normalization pro- cedure. Both of these issues need to be addressed in the fu- ture through more detailed studies on a larger sample and methods with more discriminating power (not just bulk geo- chemical data), as well as more data from the eastern Adri- atic coastal sediments. The ratios from the Palagruža core show a tendency to- wards the Apennine ratios indicating a possible influence from the WAC (Figure 9). The mean Ca/Al ratios (Supple- mentary Table 3) along the AEC indicates an increase of cal- cite/carbonate from Albania (0.77) along the eastern side of the Northern Adriatic (4.42) and then a decrease along the WAC (2.26) and the Gargano peninsula and Gulf of Manfre- Figure 9: The relationship between Ni/Al and Cr/Al ratios. The ALBANIA, SA PIT, PALAGRUŽA, JABUKA cores are from this study. Data from DOLENEC et al. (1998) includes the SAS-South Adriatic sediments, CAS-Central Adiatic sediments; NAS-North Adriatic sediments. The data shown for the Gargano peninsula, Gulf of Manfredonia, Strait of Otranto, Po di Goro and the Apenninic Rivers are from GOUDEAU et al. (2013). Geologia Croatica 67/3Geologia Croatica 202 donia have Ca/Al ratios (average 1.65 and 1.87) similar to the SA PIT (1.76) cores. The ratios for “pollution elements”, Pb/Al, Hg/Al and Zn/Al (Supplementary Table 3, Figure 8) were only available for the data from DOLENEC et al., (1998) and show that the lowest values are in the southern and central part of the east- ern Adriatic and that high values characterise the northern Adriatic sediments. The down-core distribution of estimated EFs for Hg, Cd, Zn, Cu, Pb, Ni and Cr are given in Figure 10. The calculated EFs for the top interval (0–2 cm) of each core are given in Table 3. The estimated EFs in the sediments show that only Hg can be considered as a moderate contaminant in the ana- lyzed cores. Other elements have EF>2 but based on the classification by SUTHERLAND (2000) contaminating en- richment of trace metals has not occurred. Generally the es- timated surface enrichment follows the order: Hg>Pb>Cd, the estimated EFs for Zn, Ni and Cr do not indicate any de- gree of enrichment. Mercury has the highest calculated EF of 3.6 in the Albania core while the Palagruža core has the lowest one (1.4). DOLENEC et al. (1998) concluded that most of the Hg has a source in the northern Adriatic with minor influences from the eastern coast of the Adriatic. The Figure 10: Enrichment factors (EF) estimated for Pb, Hg, Zn, Cd, Ni and Cr. Ilijanić et al.: Metal deposition in deep sediments from the Central and South Adriatic Sea Geologia Croatica 203 possible contribution from the eastern coast was detected by CUCULIĆ et al. (2009) who measured 10 times higher than average Hg concentrations in Malo Jezero on Mljet Island (located along the rim of South Adriatic Pit) during a heavy rainfall event with strong easterly winds. 6. CONCLUSIONS This study reports the concentrations of the eight trace ele- ments usually identified as priority contaminants in aquatic systems As, Cd, Cr, Cu, Hg, Ni, Pb and Zn as well as Mo, Mn, Al, Ti and Ca as lithogenic elements, in 5 short marine sediment cores, collected at locations on a transect along the western rim of the southern part of the Eastern Adriatic Cur- rent (EAC). The sediment accumulation rate for the Palagruža Sill and the South Adriatic Pit is ~0.18 cmy–1 and the top 10 cm interval of the sediment core represents approximately 50 years and a 40 cm core could have an equivalent to 200 years of deposition. The Jabuka and Albania cores presumably span shorter times (approximately 130 and 100 years, re- spectively). The sediments are highly heterogeneous and consist of carbonate and detrital aluminosilicates. The main mineral phase is calcite, followed by quartz, feldspars, micas and clay minerals (smectite, chlorite, illite and kaolinite). The trace metal down-core distributions and relations with mineral and clay composition allowed the distinction of several groups of trace metals. The “pollution elements” Pb, and Hg showed increased concentrations in the top sec- tions of the cores. Concentrations of redox-sensitive metals, Mn and Mo have vertical distributions that followed similar patterns and were used to identify the redox state of sediments. The down- core distribution of redox-sensitive trace metals are complex in the South Adriatic Pit core. The metal/Al ratios were applied to determine metal ex- cesses or to account for natural changes. They were used to give an insight to the changes in trace metals along transect of the Eastern Adriatic Current (EAC) from Albania, along the Croatian coastal islands through the Northern Adriatic, and then along the Western Adriatic Current (WAC) to the south. The Cr/Al and Ni/Al ratios used for provenance dis- crimination have ratios that are significantly different (higher) in the southern and central Adriatic along the EAC compared to those of the surface sediments of the Northern Adriatic and along the WAC (lower). The plotted Cr/Al vs Ni/Al ratios for the analysed cores, along with the available literature data (DOLENEC at al., 1998; GOUDEAU et al., 2013) show that a linear dilution path between Albania (along the AEC) and the Apennine (WAC) signature is distorted. This could indicate the poten- tial influence of another source that is decreasing the ratios, possibly from the Dinaric/Dalmatian source of minerals. The distortion could also be attributed to preferential changes in grain size towards the finer fractions which were not com- pensated by the normalization procedure. Both of these is- sues need to be addressed in the future through more detailed studies on a larger sample and methods with more discrimi- nating power (not just bulk geochemical data), as well as more data from the eastern Adriatic coastal sediments. Generally, the surface estimated enrichment follows the order: Hg>Pb>Cd, the estimated EFs for Zn, Ni and Cr do not indicate any enrichment. The estimated EFs in the sedi- ments show that only Hg can be considered as a contaminant in the analyzed sediments. ACKNOWLEDGEMENT This study is a part of several research projects, ‘Radio- ecology of the Adriatic Sea and Coastal Areas’, ‘Environ- mental Radioactivity and Radiation Protection’ and “The Ba- sic Geochemical Map of the Republic of Croatia” supported by the Croatian Ministry of Science, Education and Sports of the Republic of Croatia and IAEA TC project RER/7/003 ‘Marine Environmental Assessment of the Mediterranean Region’. REFERENCES ACQUAVITA, A., COVELLI, S., EMILI, A., BERTO, D., FAGANELI, J., GIANI, M., HORVAT, M., KORON, N. & RAMPAZZO, F. (2012): Mercury in the sediments of the Marano and Grado Lagoon (northern Adriatic Sea): Sources, distribution and speciation.– Es- tuar. Coast. Shelf. Sci., 113, 20–31. AMOROSI, A. (2012): Chromium and nickel as indicators of source-to- sink sediment transfer in a Holocene alluvial and coastal system (Po Plain, Italy).– Sedim. Geol., 280, 260–269. AMOROSI, A., CENTINEO, M.C., DINELLI, E., LUCCHINI, F. & TATEO, F. (2002): Geochemical and mineralogical variations as indicators of provenance in late Quaternary deposits of SE Po Plain.– Sedim. Geol., 151, 273–292. ANNIBALDI, A., TRUZZI, C., ILLUMINATI, S. & SCARPONI, G. (2009): Recent sudden decrease of lead in Adriatic coastal seawater during the years 2000–2004 in parallel with the phasing out of lead- ed gasoline in Italy.– Mar. Chem., 113, 238–249. ANSCHUTZ, P., DEDIEU, K., DESMAZES, F. & CHAILLOU, G. (2005): Speciation, oxidation state, and reactivity of particulate manganese in marine sediments.– Chem. Geol., 218/3–4, 265–279. BECK, M., DELLWIG, O., SCHNETGER, B. & BRUMSACK, H.J (2008): Cycling of trace metals (Mn, Fe, Mo, U, V, Cr) in deep pore waters of intertidal flat sediments.– Geoch. Cosmoch. Acta, 72/12, 2822–2840. BOGNER, D., JURAČIĆ, M., ODŽAK, N. & BARIĆ, A. (1998): Trace metals in fine grained sediments of the Kaštela bay, Adriatic Sea.– Wat. Sci. Techn., 38/11, 169–175. BULJAN, M. & ZORE-ARMANDA, M. (1976): Oceanographical prop- erties of the Adriatic Sea.– Ocean. Mar. Biol. Ann. Rev., 14, 11–98. CALANCHI, N., DINELLI, E., LUCCHINI, F. & MORDENTI, A. (1996): Chemostratigraphy of late Quaternary sediments from Lake Albano and central Adriatic Sea cores (PALICLAS Project).– Mem. Ist. ital. Idrobiol., 55, 247–263. CHAILLOU, G., SCHÄFER, J., BLANC, G. & ANSCHUTZ, P. (2008): Mobility of Mo, U, As, and Sb within modern turbidites.– Mar. Ge- ol., 254/3–4, 171–179. COVELLI, S., FAGANELI, J., HORVAT, M. & BRAMBATI, A. (2001): Mercury contamination of coastal sediments as the result of long- term cinnabar mining activity (Gulf of Trieste, northern Adriatic sea).– Appl. Geoch., 16, 541–558. Geologia Croatica 67/3Geologia Croatica 204 COVELLI, S., FONTOLAN, G., FAGANELI, J. & OGRINC, N. (2006): Anthropogenic markers in the Holocene stratigraphic sequence of the Gulf of Trieste (northern Adriatic Sea).– Mar. Geol., 230, 29–51. COVELLI, S., PIANI, R., ACQUAVITA, A., PREDONZANI, S. & FA- GANELI, J. (2007): Transport and dispersion of particulate Hg as- sociated with a river plume in coastal Northern Adriatic environ- ments.– Mar. Poll. Bull., 55, 436–450. CUCULIĆ, V., CUKROV, N., KWOKAL, Ž. & MLAKAR, M. (2009): Natural and anthropogenic sources of Hg, cd, Pb, Cu and Zn in sea- water and sediment of Mljet Natinal Park, Croatia.– Est. Coast. Shelf Sci., 81, 311–320. CUKROV, N., FRANČIŠKOVIĆ-BILINSKI, S., HLAČA, B. & BA- RIŠIĆ, D. (2011): A recent history of metal accumulation in the sediments of Rijeka harbor, Adriatic Sea, Croatia.– Mar. Poll. Bull., 62/1, 154–167. DE LAZZARI, A., RAMPAZZO, G. & PAVONI, B. (2004): Geochemi- stry of sediments in the Northern and Central Adriatic Sea.– Est. Coast. Shelf Sci., 59, 429–440. DI LEONARDO, R., TRANCHIDA, G., BELLANCA, A., NERI, R., ANGELONE, M & MAZZOLA, S. (2006): Mercury levels in sed- iments of central Mediterranean Sea: A 150+ year record from box- cores recovered in the Strait of Sicily.– Environ. Chem., 65/11, 2366–2376. DINELLI, E. & LUCCHINI, F. (1999): Sediment supply to the Adriatic sea basin from the Italian rivers: geochemical features and environ- mental constraints.– Giorn. Geol., 61, 121–132. DINELLI, E., TATEO, F. & SUMMA, V. (2007): Geochemical and min- eralogical proxies for grain size in mudstones and siltstones from the Pleistocene and Holocene of the Po River Alluvial Plain, Italy.– Geolog. Soc. Am. Spec. Pap., 420, 25–36. DOLENEC, T. (2003): Ferromanganese coated structures from the Jabu- ka Pit (Central Adriatic): Mineralogical, geochemical and genetic considerations.– Croat. Chem. Acta, 76/3, 201–215. DOLENEC, T., FAGANELI, J. & PIRC, S. (1998): Major, minor and tra ce elements in surficial sediments from the open Adriatic Sea: A regional geochemical study.– Geol. Croat., 51/1, 59–73. EBERL, D.D. (2003): User guide to RockJock – a program for determin- ing quantitative mineralogy from X-ray diffraction data.– US Geol. Surv. Open File Rep., 03–78. EICHLER, A., TOBLER, L., EYRIKH, S., GRAMLICH, G., MALY- GINA, N., PAPINA, T. & SCHWIKOWSKI, M. (2012): Three Centuries of Eastern European and Altai Lead Emissions recorded in a Belukha Ice Core. Environ.– Sci. Technol., 46, 4323−4330. EMILI, A., ACQUAVITA, A., KORON, N., COVELLI, S., FAGANELI, J., HORVAT, M., ŽIŽEK, S. & FAJON, V. (2012): Benthic flux measurements of Hg species in a northern Adriatic lagoon environ- ment (Marano and Grado Lagoon, Italy).– Est. Coast. Shelf Sci., 113, 71–84. FAGANELI, J., HORVAT, M., COVELLI, S., FAJON, V., LOGAR, M., LIPEJ, L. & CERMELJ, B. (2003): Mercury methylmercury in the Gulf of Trieste (northern Adriatic Sea).– Sci. Tot. Env., 304, 315–326. FERGUSON, J.E. (1990): The heavy metals: chemistry, environmental impact and health effects.– Pergamon, New York, 412 p. FÖRSTNER, U. & WITTMANN, G.T.W. (1981): Metal pollution in the aquatic environment.– Springer, Heidelberg, 486 p. FRIGNANI, M., SORGENTE, D., LANGONE, L., ALBERTAZZI, S. & RAVAIOLI, M. (2004): Behavior of Chernobyl radiocesium in sediments of the Adriatic Sea off the Po River delta and the Emilia- Romagna coast.– J. Environ. Radioactiv., 71, 299–312. FRIGNANI, M., LANGONE, L., RAVAIOLI, M., SORGENTE, D., ALVISI, F. & ALBERTAZZI, S. (2005): Fine-sediment mass bal- ance in the western Adriatic contitnental shelf over a century time scale.– Mar. Geol., 222–223, 113–133. GARCIA, D., RAVENNE, C., MARÉCHAL, B. & MOUTTE, J. (2004): Geochemical variability induced by entrainment sorting: quantified signals for provenance analysis.– Sed. Geol., 171, 113–128. GIANI, M., DJAKOVAC, T., DEGOBBIS, D., COZZI, S., SOLIDORO, C. & FONDA UMANI, S. (2012): Recent changes in the marine ecosystems of the northern Adriatic Sea.– Est. Coast. Shelf Sci., 115, 1–13. GOUDEAU, M.L.S., GRAUEL, A.L. & BERNASCONI, S.M. (2013): Provenance of surface sediments along the southeastern Adriatic coast off Italy: An overview.– Est. Coast. Sh. Sci., 134, 45–56. HALAMIĆ, J., PEH, Z., MIKO, S., GALOVIĆ, L. & ŠORŠA, A. (2012): Geochemical Atlas of Croatia: Environmental Implications and Ge- ochemical Thread. – J. Geoch. Explor., 115, 36–46. LEDER, N. (2004): Adriatic Sea Pilot. Part B-1. Hydrographic Institute of the Republic of Croatia. LI, Y.H. (1981): Geochemical cycles of elements and human perturba- tion.– Geoch. Cosmoch. Acta, 45, 2073–2084. LOVRENČIĆ MIKELIĆ, I., OREŠČANIN, V. & BARIŠIĆ, D. (2013): Distribution and origin of major, minor, and trace elements in sed- iments and sedimentary rocks of the Kaštela Bay (Croatia) coastal area.– J. Geoch. Expl., 128, 1–13. LUCCHINI, F., DINELLI, E. & MORDENTI, A. (2003): Geochemical records of palaeoenvironmental changes during Late Quaternary in the Adriatic Sea sediments.– GeoActa, 2, 43–62. MASON, R.P., CHOI, A.L., FITZGERALD, W.F., HAMMER- SCHMIDT, C.R., LAMBORG, C.H., SOERENSEN, A.L. & SUN- DERLAND, E.M. (2012): Mercury biogeochemical cycling in the ocean and policy implications.– Environ. Res., 119, 101–117. MIKO, S., HALAMIĆ, J., PEH, Z. & GALOVIĆ, L. (2001): Geochem- ical Baseline Mapping of Soils Developed on Diverse Bedrock from Two Regions in Croatia.– Geologia Croatica, 54/1, 53–118. MIKO, S., KOCH, G., MESIĆ, S., ŠPARICA MIKO, M., ŠPARICA, M., VREČA, P. & DOLENEC, T. (2007): Influence of land use in small karst watersheds on chemical status of peloid sediments on the Eastern Adriatic Coast.– J. Soils Sed., 7/5, 303–312. MOORE, D.M. & REYNOLDS, R.C. (1997): X-ray diffraction and the identification and analysis of clay minerals.– Oxford Univ. Press, Oxford, New York, 332 p. MORFORD, J.L. & EMERSON, S. (1999): The geochemistry of redox sensitive trace metals in sediments.– Geoch. Cosmoch. Acta, 63/11– 12, 1735–1750. NAMEROFF, T. J., BALISTRIERI, L.S., MURRAY J. W. (2002): Sub- oxic trace metal geochemistry in the eastern tropical North Pacific.– Geochimica et Cosmochimica Acta, 66/7, 1139–1158. NELSON, B.W. (1972): Mineralogical differentiation of sediments dis- persed from the Po delta.– In: STANLEY, D.J. (ed.): The Mediter- ranean Sea: a natural sedimentation laboratory, Dowden, Hutchin- son & Ross, Stroudsburg, 441–453. OBHOĐAŠ, J., VALKOVIĆ, V., MATJAČIĆ, L., NAĐ, K. & SUDAC, D. (2012): Evaluation of elemental composition of sediments from the Adriatic Sea by using EDXRF technique.– Appl. Rad. Isot., 70/7, 1392–1395. ORLIĆ, M., GAČIĆ, M. & LA VIOLETTE, P.E. (1992): The currents and circulation of the Adriatic Sea.– Ocean. Acta, 15/2, 109–124. OTHMAN, I., AL-MASRI, M.S. & AL RAYYES, A.H. (2000): Sedi- mentation rates and pollution history of the eastern Mediterranean Sea: Syrian coast.– Sci. Tot. Envir., 248, 27–35. PALINKAS, C.M. & NITTROUER, C.A. (2007): Modern sediment ac- cumulation on the Po shelf, Adriatic Sea – Cont. Shelf. Resear., 27, 489–505. PETRINEC, B., FRANIĆ, Z, ILIJANIĆ, N., MIKO, S., ŠTROK, M. & SMODIŠ, B. (2012): Estimation of sedimentation rate in the mid- dle and south Adriatic Sea using 137Cs.– Radiat. Protect. Dos., 151/1, 102–111. Ilijanić et al.: Metal deposition in deep sediments from the Central and South Adriatic Sea Geologia Croatica 205 PETRINEC, B., ŠTROK, M., FRANIĆ, Z., SMODIŠ, B. & PAVIČIĆ- HAMER, D. (2013): Radionuclides in the Adriatic sea and related dose-rate assessment for marine biota.– Radiat. Protect. Dos., 154/3, 320–330. PIGORINI, B. (1968): Sources and dispersion of recent sediments of the Adriatic Sea.– Mar. Geol., 6, 187–229. PROHIĆ, E. & KNIEWALD, G. (1987): Heavy metal distribution in re- cent sediments of the Krka River Estuary – an example of sequen- tial extraction analysis.– Mar. Chem., 22/2–4, 279–297. RAJAR, R., ČETINA, M., HORVAT, M. & ŽAGAR, D. (2007): Mass balance of mercury in the Mediterranean Sea.– Mar. Chem., 107/1, 89–102. REIMANN, C. & DE CARITAT, P. (2005): Distinguishing between nat- ural and anthropogenic sources for elements in the environment: regional geochemical surveys versus enrichment factors.– Sci. Tot. Envir., 337, 91–107. RIDGWAY, J. & SHIMMIELD, G. (2002): Estuaries as repositories of historical contamination and their impact on shelf seas.– Estuar. Coast. Shelf. Sci., 55, 903–92. ROMANO, S., LANGONE, L., FRIGNANI, M., ALBERTAZZI, S., FOCACCIA, P., BELLUCCI, L.G. & RAVAIOLI, M. (2013): His- torical pattern and mass balance of trace metals in sediments of the northwestern Adriatic Shelf.– Mar. Poll. Bull., 76, 32–41. SACCANI, E. & PHOTIADES, A. (2005): Petrogenesis and tecton- omagmatic significance of volcanic and subvolcanic rocks in the Albanide–Hellenide ophiolitic mélanges.– The Island Arc, 14, 494–516. SHOTYK, W., WEISS, D. & KRAMERS, J.D. (2001): Geochemistry of the peat bog at Etang de la Grue`re, Jura Mountains, Switzerland, and its record of atmospheric Pb and lithogenic trace elements (Sc, Ti, Y, Zr, Hf and REE) since 12,370 14C yr BP.– Geoch. Cosmoch. Acta, 65/14, 2337–2360. SONDI, I., LOJEN, S., JURAČIĆ, M. & PROHIĆ, E. (2008): Mecha- nisms of land–sea interactions – the distribution of metals and sed- imentary organic matter in sediments of a river-dominated Mediter- ranean karstic estuary.– Estuar. Coast. Shelf. Sci., 80/1, 12–20. SPAGNOLI, F., BARTHOLINI, G., DINELLI, E. & GIORDANO, P. (2008): Geochemistry and particles size of surface sediments of Gulf of Manfredonia (Southern Adriatic Sea).– Estuar. Coast. Sh. Sci., 80, 21–30. SPAGNOLI, F., DELL’ANNO, A., DE MARCO, A., DINELLI, E., FA- BIANO, M., GADALETA, M.V. & IANNI, C. (2010): Biogeo- chemistry, grain size and mineralogy of the central and southern Adriatic Sea sediments: a review.– Chem. Ecol., 26, 19–44. STATSOFT, INC. (2006): STATISTICA (data analysis software system), version 7.1, Tulsa, SAD. SUTHERLAND, R.A. (2000): Bed sediment-associated trace metals in an urban stream, Oahu, Hawaii. – Environ. Geol., 39/6, 611–627. ŠTROK, M., SMODIŠ, B. & PETRINEC, B. (2010): Natural radionu- clides in sediments and rocks from Adriatic Sea. – J. Radianal. Nu- cl. Chem., 286, 303–308. TANKÈRÉ, S.P.C., PRICE, N.B. & STATHAM, P.J. (2000): Mass bal- ance of trace metals in the Adriatic Sea.– J. Mar. Syst., 25, 269–286. TOMADIN, L. (2000): Sedimentary fluxes and different dispersion mechanisms of the clay sediments in the Adriatic Basin.– Rend. Lincei, 9/11, 161–174. VAN DER WEIJDEN, C.H. (2002): Pitfalls of normalization of marine geochemical data using a common divisor.– Mar. Geol., 184/3–4, 167–187. VITAL, H. & STATTEGGER, K. (2000): Major and trace elements of stream sediments from the lowermost Amazon River.– Chem. Ge- ol., 168, 151–168. WELTJE, G.J. & BROMMER, B. (2011): Sediment-budget modelling of multi-sourced basin fills: application to recent deposits of the western Adriatic mud wedge (Italy).– Basin Resear., 23, 291–238. WINDOM, H., SCHROPP, S., CALDER, F., RYAN, J., SMITH, R.J., BURNEY, L. LEWIS, F. & RAWLISON, C. (1989): Natural trace metal concentrations in estuarine and coastal marine sediments of the southeastern U.S. – Environ. Sci. Technol., 23/3, 314–320. ZORE ARMANDA, M. & GAČIĆ, M. (1987): Effects of the Bura on the circulation in the North Adriatic.– Ann. Geophy, 5B, 93–102. Manuscript received January 14, 2014 Revised manuscript accepted September 25, 2014 Available online October 31, 2014