Sediment geochemistry and accumulation rates on the northeastern shelf of the Gulf of Cádiz ( SW Iberian Peninsula )

1 Centro Interdipartimentale di Ricerca per le Scienze Ambientali (C.I.R.S.A.), University of Bologna, Via S.Alberto 163, Ravenna 48100, Italy. E-mail: roberta.guerra@unibo.it 2 Department of Physics, University of Bologna, Viale B. Pichat 6/2, Bologna 40126, Italy. 3 Department of Earth and Geological-Environmental Sciences, University of Bologna, Piazza Porta San Donato 1, Bologna 40127, Italy. 4 Departamento de Química Física, Facultad Ciencias del Mar y Ambientales (CASEM), Polígono Río San Pedro s/n, 11510 Puerto Real, Cádiz, Spain.


INTRODUCTION
The importance of the Gulf of Cadiz continental margin (SW Spain) lies in its proximity to the Strait of Gibraltar, a point of singular geographic interest because it is the place where the Atlantic Ocean and the Mediterranean Sea interchange their water masses (Gutiérrez-Mas et al., 2003).The configuration of the study area in this study is the result of a natural evolution determined by geological agents, principally the littoral dynamics, the mouth of the river Gua dalquivir and the aeolian regime of the zone (Ligero et al., 2005).Continental shelves (defined as 30-200 m areas) represent <10% of the global oceanic area (Berelson et al., 2003) and are considered as the transition zone between the continents and the open ocean.They receive large amounts of terrestrial organic matter and nutrients, mainly through river systems.Because the mean water depth of continental shelves is significantly less than in the open ocean, the seafloor sediments play a significantly more important role in the biogeochemistry and ecology of these systems (MacKenzie et al., 2005).
Other major topics that geochemical studies can face in these areas are those of sediment provenance, which is useful to define sediment sources and their dispersion patterns, and to assess local human impacts (Carranza-Edwards et al., 2005;Spagnoli et al., 2008).
Natural and artificial radionuclides such as 210 Pb and 137 Cs can provide useful information as tracers to determine time scales for various sedimentary processes affecting continental shelves (Sánchez-Cabeza et al., 1999;Palinkas and Nittrouer, 2007;Jouanneau et al., 2008). 210Pb (half life: 22.3 years) is a naturally occurring radionuclide belonging to the 238 U series whereas 137 Cs (half life: 30.2 years) is produced artificially as a result of anthropogenic activities.
Two cruises were carried out in two periods of the year 2006 on the northeastern shelf of the Gulf of Cádiz.This paper presents the results obtained from the study of sediment cores that were sampled on these cruises in order to characterize in this area the geochemistry of sediments and the accumulation rates by means of 210 Pb and 137 Cs radioisotopes.

Study site
The study was carried out on the northeastern shelf of the Gulf of Cádiz (southwestern coast of the Iberian Peninsula; Fig. 1).The Gulf of Cádiz is the meeting place for the North Atlantic Ocean and the Mediterranean Sea through the Strait of Gibraltar.This gulf is a wide basin influenced by: a) the dense plume of Mediterranean water that penetrates through the strait, and b) the freshwater inputs from several adjacent rivers, such as the Guadiana, the Guadalquivir, the Tinto and the Odiel.
The River Guadalquivir is the main fluvial source draining into the Gulf of Cádiz margin, with a mean annual water discharge of 160 m 3 s -1 (Van Geen et al., 1997).The circulation on the northeastern shelf of the Gulf of Cádiz is mainly controlled by: a) North Atlantic Surface Water (NASW), which flows towards the east and southeast to the Strait of Gibraltar, and b) an intermittent counter-current system which seems to be considerably induced by wind forcing (Lobo et al., 2004).The Cádiz area has a regime of semidiurnal tides and is characterized by predominant winds from the west and the east.
The data reported in this paper belong to two cruises on board the R/V Mytilus covering two seasonal periods: June 2006 and November 2006 (summer and autumn, respectively).Three sampling stations were selected for the study, and sediment cores were sampled by scuba divers.Two stations were located in the Bay of Cádiz and its neighbouring outer region (BC4 and BC5) and the other one off the mouth of the River Guadalquivir (GL3) (Fig. 1).Station GL3 is located on the Guadalquivir submarine prodelta, which is characterized by high fluvial supply and a moderate hydrodynamic regime (Lobo et al., 2004), whereas the BC stations are located in a more energetic area affected by the tidal currents between the Bay of Cádiz and the outer shelf.Sediments in the study area are mainly siliciclastic, with 25% of bioclastic carbonates, predominantly calcite (Gutiérrez-Mas et al., 2003).
Specifically, in this research work three cores were analyzed: GL3 (from station GL3), which was extracted in June 2006, and BC4 and BC5 (from stations BC4 and BC5), which were extracted in June 2006 and November 2006, respectively.
Cores were extracted without alteration to the structure of the sediment.In order to ensure the immobilization of the core, the pipes were cooled at -10ºC.Afterwards, the sediment profiles were cut into 1-cm thick slices up to a depth of 14-15 cm for GL3, 9-10 cm for BC4 and 13-14 cm for BC5.

Organic carbon and total nitrogen analysis
Sediment samples were ground to a fine powder with an agate pestle and mortar.Total carbon (TC) and nitrogen (TN) contents were determined using elemental analysis of combusted aliquots (30 mg) with a TCM 480 Carlo Erba Elemental Analyzer.Total organic carbon (TOC) was measured on decarbonated samples (1.5 N HCl) following the method of Hedges and Stern (1984).The precision of this method based on replicate measurements (n=30) of a standard reference sediments BCSS-1 is 2.0% for TOC and 1.2% for TN.

Sediment geochemistry
Bulk chemical composition for major and minor elements (Si, Ti, Al, Fe, Mg, Ca, K, Na, P, S) and trace elements (Mn, Sc, V, Cr, Ni, Cu, Zn, Rb, Sr, Y, Zr, Ba, Ce, Pb) were determined by X-ray fluorescence spectrometry using a Philips PW 1480 spectrometer equipped with an Rh tube on pressed powder pellets following the matrix correction methods of Franzini et al. (1972) and Leoni et al. (1982).About 30 international reference materials were used for the calibration, and PACS-2 (marine sediment, National Research Council, Canada), NBS1646a (estuarine sediment, National Institute of Standards and Technology, USA) as external monitors.Precision and accuracy were better than 5% for trace-element determinations.
Total mercury was determined by cold vapour atomic absorption spectrometry (CVAAS) after hot digestion of 0.5-1 g dry sediment with concentrated HNO 3 + H 2 SO 4 for 3 h at 90°C (Fabbri et al., 2001).The analytical precision of the Hg data, estimated as relative percent difference among replicates, was 5.0% (n=10), and analytical accuracy estimated as the mean% recovery in PACS-2 reference sediment (National Research Council, Canada) was 95% (n=10).

Radioactive tracer analyses
Water content and wet and dry bulk densities were calculated from samples that were weighed before and after drying at 60°C until constant weight.Dry homogenized samples were packed in 9.5-mL sealed Petri dishes and 'aged' at least four weeks to allow the equilibrium between 222 Rn and its descendants, 214 Bi and 214 Pb, to be established for the determination of 226 Ra.Radioactive tracers ( 210 Pb, 226 Ra, and 137 Cs) were measured on sediment core BC4 using HPGe coaxial detector with an energy resolution (full-width at half-maximum) of 1.9 keV at 1.33 MeV ( 60 Co), and a 22.6% efficiency.Efficiency calibration was performed using a Standard U-Th Ore (Canada Centre for Mineral and Energy Technology).Reference sediments IAEA 315 and IAEA 368 from the International Atomic and Energy Agency (IAEA) were used to check the accu-racy of the results.Samples and background were counted for nominally 172800 s (corresponding to two days). 210Pb activity was determined by direct measurements of its gamma decay energy at 46.5 keV.Activity concentrations of 214 Pb and 214 Bi were quantified using the 351.9 keV and 609.3 keV photopeaks, respectively. 137Cs (half life: 30.2 years) was measured by its emissions at 662 keV.Excess 210 Pb ( 210 Pb xs ) activity was calculated by subtracting the 226 Ra activity from the total 210 Pb.

Sediment accumulation rate
Sediment reworking due to physical or biological agents may affect the excess 210 Pb distribution along the sediment.Goldberg and Koide (1962) pointed out that the main processes governing excess 210 Pb profiles in the seabed are sediment accumulation rates, radioactive decay and particle mixing.They proposed a onedimensional advection-diffusion model to calculate the sedimentation rate (S, in cm yr -1 ) and the mixing coefficient (D b ; in cm 2 yr -1 ) that describes the intensity of particle reworking (D b , in cm 2 yr -1 ): where A is the excess 210 Pb activity (Bq kg -1 ) at a depth x (cm), l is the 210 Pb decay constant (0.0311 yr -1 ), and S and D b are assumed to be constant.As D b and S cannot be determined independently, a solution for D b can be obtained if S is known or assumed to be negligible.When the latter is not the case or no evidence supporting this is available, S can be determined from the 210 Pb profile of the non-mixed layer, where D B is assumed negligible.Assuming steady state conditions and no mixing, Equation (1) can be solved under the boundary conditions of A = A 0 (x = 0) and A→ 0 (x → ∞) by means of the equation: This is usually done by least-squares fitting of the logarithm of excess 210 Pb versus depth for the strata below the sediment mixed layer (SML).Then, the sedimentation rate calculated by using Equation (2) can be introduced as a constant in Equation (1) to determine D B , also using least-square fitting for the SML: In this study, we consider the 210 Pb profile of core BC4 from the outer Bay of Cádiz as a two-layer system with an upper mixed layer extending to a distance L below the water-sediment interface (SML) and a second layer below L where no mixing takes place (Fig. 5).

Sediment geochemistry
The vertical profiles of major and minor elements (Si, Ti, Al, Fe, Mg, Ca, K, Na, P, S) and selected trace elements (Mn, Sc, V, Cr, Ni, Cu, Zn, Rb, Sr, Y, Zr, Ba, Ce, Pb, Hg) are reported in Figures 3 and 4 respectively.The profiles of the three cores are reported in the same diagram to facilitate comparison and discussion of the most relevant features.
Trace metals of environmental interest (Fig. 4 and Table 1) did not reach a high maximum concentration (Table 1) and displayed rather constant downcore profiles both in the outer Bay of Cádiz and in the Guadalquivir submarine prodelta area (Fig. 4), except for Hg in core BC4 and BC5, where an enrichment at the top was observed.The concentrations of trace metals in core GL3 were higher than those in the two cores of the Bay of Cádiz.

Pb and 137 Cs profiles and sediment accumulation rates
Data of bulk density and excess 210 Pb activity measured in the outer Bay of Cádiz (core BC4) are reported in Table 2; the 210 Pb profile is shown in Figure 5.The excess 210 Pb activity varied from about 40-50 Bq kg -1 in the upper layers to 5 Bq kg -1 in the deeper layer.On the basis of the excess 210 Pb activity profile, -Depth profiles of trace elements in the sediment cores from the Gulf of Cádiz.The origin of the cores is shown in Figure 1.Sediment cores BC4 and BC5 were obtained from the outer Bay of Cádiz, whereas core GL3 was obtained from the submarine prodelta of the River Guadalquivir.
we considered the surface mixed layer (SML) to be the slice from 0 to 5 cm and the non-mixed layer to be the slice from 5 to 10 cm.When applying Equation 2to the non-mixed layer of core BC4, a sedimentation rate S of 0.10 ± 0.01 cm yr -1 (r = 0.9847, p = 0.0007) and an apparent sediment accumulation rate (SAR) of 0.11 ± 0.01 g cm -2 yr -1 (r = 0.9867, p = 0.0005) were obtained.The 210 Pb mixing-coefficient (D b ) calculated with Equation 3 was 1.07 ± 0.13 cm 2 yr -1 (r = 0.77, p = 0.00003).

Total organic carbon (TOC) and total nitrogen (TN) distribution and ratios
TOC values found in the sediments under the influence of the River Guadalquivir (core GL3) and those from the Bay of Cádiz (core BC4 and BC5) fall within the typical low range for coastal sediments (Dickens et al., 2004), and are close to those reported for marine sediments from the inner continental shelf of the southwest Atlantic coast of Spain (López-Capel et al., 2006;Sánchez-García et al., 2008).
The most diagnostic point is the C/N index (atomic TOC/TN ratio), which can be used to distinguish among potential organic matter origins (e.g.marine phytoplankton C/N = 7.4 ± 1.3, and cellulose rich vascular land plants C/N≥ 20; Meyers, 1994Meyers, , 1997;;Anderson and Sarmiento, 1994).Thus, the C/N values found here (Fig. 2), mostly around 10 with the exception of few layers in core GL3, seem to be attributable to a predominantly marine origin for the sedimentary organic matter (OM).However, one should be cautious with such an interpretation, since the observed d 15 N values in the top 20 cm from the Gulf of Cádiz (2.9-6.1‰,Sánchez- García et al., 2008) are likely to indicate a mixed contribution of both marine and terrestrial organic matter.

Sediment geochemistry
Core GL3 from the area under the influence of the River Guadalquivir has a completely different composition to that of the other two cores.It has higher concentrations of Al 2 O 3 and other elements such as Fe 2 O 3 , MgO, K 2 O, V, Ni, Cu, Zn, Rb, Pb (Figs. 3 and 4) that can be related to a clay mineral fraction, which is dominant in the sediments of this area (López-Galindo et al., of these elements in core GL3 are rather regular, with a minimum in concentration for almost all elements at 7 cm depth within the core.These elements have lower concentrations and generally comparable downcore profiles in the two cores (BC4 and BC5) from the outer Bay of Cádiz (Figs. 3 and 4).Some difference is evident for K 2 O, and in general, core BC5 displays a greater variation in the topmost 5 cm, with an increase between 2 and 3 centimetre followed by a more or less regular trend.
SiO 2 concentrations are high in cores from the outer Bay of Cádiz, and display profiles that are generally opposite to the clay mineral fraction described above (Fig. 3).The SiO 2 /Al 2 O 3 ratio shows a strong positive correlation with grain-size in several cases (Viscosi-Shirley et al., 2003;Dinelli et al., 2007) and in the cores of the Bay of Cádiz the values are typical of a sandy sediment (5.5-7.2SiO 2 /Al 2 O 3 in BC4, 5.5-8.4SiO 2 /Al 2 O 3 in BC5), whereas in core GL3 the lower values (3.4-4.2) point to a consistently finer mean grain-size, in accordance with the grain-size data pub-  lished by López-Galindo et al. (1999) and Gutiérrez-Mas et al. (2003).The significant enrichment of Y and Zr in both cores from the outer Bay of Cádiz can be related to sorting effects on trace minerals (e.g.zircon) (Fig. 4).A similar effect is common in sedimentary environments (Fralick and Kronberg, 1997;Dypvik and Harris, 2001;García et al., 2004), and is related to the action of bottom currents that selectively remove the fine grain size fraction and concentrate the heavy minerals.In general, trace metals do not reach very high concentrations, except Pb in core GL3, which reaches maximum concentrations of 65 mg kg -1 (Fig. 4), three times the average shale composition (20 mg kg -1 , Turekian and Wedephol, 1961).However, this result is consistent with data reported for the Guadalquivir estuary (Riba et al., 2004).
The vertical profiles of the trace metals do not indicate surface enrichment in sediments from the outer Bay of Cádiz; however, a slight increase in Cr, Cu, Zn and Pb concentrations in core BC5 can be observed in the topmost 8 cm (Fig. 4), and the same feature can also be observed in BC4, though only one sample shows reduced values at the bottom of the core.No significant downcore changes can be observed in core GL3.
Cores collected in the outer Bay of Cádiz show similar mercury profiles (Fig. 4), with higher concentrations at the top decreasing to nearly constant levels with depth.Mercury levels found in the upper sections (133-168 ng g -1 and 119-148 ng g -1 , respectively) are consistently lower than those found in the inner Bay and littoral ecosystems in the Gulf of Cádiz (DelValls et al., 2002;Ligero et al., 2002).The lowest levels were found in deeper layers of cores BC4 and BC5 (45-49 ng g -1 and 37-64 ng g -1 , respectively), showing values comparable to those detected in mud volcano cores collected from the Gulf of Cádiz (Mieiro et al., 2007).Mercury concentrations of about 50 ng g -1 were also observed at the base of 2-m cores collected in the Gulf of Cádiz, and resulted from transport by the Rivers Tinto and Odiel (Cossa et al., 2001).Conversely, core GL3 from the Guadalquivir area exhibited the highest concentrations, with mercury peaks observed at different depths (Fig. 4).These results are consistent with the fact that this core was collected in an area located in the submarine prodelta of the River Guadalquivir, which is the main fluvial input of the Gulf of Cádiz.Hg levels found in this core fall within the range detected in sediments from the River Guadiamar, a tributary of the River Guadalquivir, and in littoral sediments under the influence of the Odiel system (Alastuey et al., 1999;Usero et al., 2005;Sáinz and Ruiz, 2006).

Pb and 137 Cs profiles and sediment accumulation rate
The magnitude of calculated sedimentation rate and sediment accumulation rate (SAR) in the outer Bay of Cádiz broadly reflects values that have also been found in shelves of the mid-Atlantic, the western Mediterranean, and the North Iberian Margin (Sánchez-Cabeza et al., 1999;Masqué et al., 2003;Sommerfield, 2006;Jouanneau et al., 2008).Bioturbation rates closely resemble mixing in deep-sea sediments rather than the much more rapid mixing seen in shallow-water environments (Schmidt et al., 2002;Masqué et al., 2003;Schmidt et al., 2007).The effects of mixing are therefore reasonably low for the Cádiz shelf environment, but should be considered for high resolution or precise geochronological studies.
137 Cs activities detected were well below 10 Bq kg -1 (Fig. 5), in agreement with the findings of Ligero et al. (2002;2005), but the 137 Cs inventory (26 mBq cm -2 ) was lower than the average inventory found in the inner Bay of Cádiz (Ligero et al., 2005).A strong positive correlation between 137 Cs and organic carbon was found (r 2 = 0.84, p = 0.0002), suggesting that the capacity of retention of 137 Cs in the sediment is linked to the organic matter (Rubio et al., 2003).However, mobility of 137 Cs in the sediment column through processes of molecular diffusion and biological activity cannot be excluded in this area, considering that BC sites showed slightly higher faunal density and biomass than GL sites, as well as greater species richness according to the findings of Ferrón et al. (2009).
The core (BC4) does not show maximum values that can be directly associated with the historical fallout of 137 Cs, in agreement with other studies carried out in Spanish coastal and marine ecosystems (Gascó et al., 1999;Rubio et al., 2003;Ligero et al., 2005).The impact of the Chernobyl disaster has not been great in Spain, as can be inferred from the data reported by UNSCEAR (1988) andDe Cort et al. (1998).Average deposition densities for 137 Cs after the Chernobyl disaster were about 0.02 kBq m -2 in the western Spanish regions, and about 0.07 kBq m -2 in the eastern regions.Finally, despite the close proximity, the Gulf of Cádiz was probably not contaminated by fallout from the accident in Algeciras in 1998; this conclusion agrees with plume trajectory simulations carried out by Vogt et al. (1998) and Quelo et al. (2007).

CONCLUSIONS
Sediment cores from the outer Bay of Cádiz display a similar composition, reflecting a relatively coarse grained texture, whereas sediments in the Guadalquivir prodelta area are dominated by finer grain sizes, as suggested from bulk geochemical analyses.Levels of trace metals (V, Ni, Cu, Zn, Pb and Hg) are higher in the Guadalquivir prodelta area than in the Bay of Cádiz area and are influenced by the contribution of the heavily contaminated sediments from the River Guadalquivir.In this area, trace metals do not vary with depth, whereas a shift in concentration is observed in the Bay of Cádiz cores.Future studies are needed to extend the application of the 210 Pb mixing model for estimating bioturbation and sediment accumulation rates to other locations in the Bay of Cádiz and to the Gualdalquivir area, where no previous data are available.

Fig. 1 .
Fig. 1. -Location of the study area and sampling area on the northern shelf of the Gulf of Cádiz (SW Spain).Sediment cores BC4 and BC5 were obtained from the outer Bay of Cádiz, whereas core GL3 was obtained from the submarine prodelta of the River Guadalquivir.

Fig. 2 .
Fig. 2. -Depth profiles of TC, OC, TN, and OC:TN molar ratios of sediment cores from the Gulf of Cádiz.The origin of the cores is shown in Figure 1.Sediment cores BC4 and BC5 were obtained from the outer Bay of Cádiz, whereas core GL3 was obtained from the submarine prodelta of the River Guadalquivir.

Table 1
. -Concentrations of metals measured in each core.Results for V, Cr, Ni, Cu, Zn and Pb are expressed in μg•g -1 , whereas results for Hg are expressed in ng•g -1 .Sediment cores BC4 and BC5 were obtained from the outer Bay of Cádiz, whereas core GL3 was obtained from the submarine prodelta of the River Guadalquivir.Fig.3.-Depthprofiles of major elements in the sediment cores from the Gulf of Cádiz.The origin of the cores is shown in Figure1.Sediment cores BC4 and BC5 were obtained from the outer Bay of Cádiz, whereas core GL3 was obtained from the submarine prodelta of the River Guadalquivir.72 • R. GUERRA et al.SCI.MAR., 74S1, December 2010, 67-76.ISSN 0214-8358 doi: 10.3989/scimar.2010.74s1067

Table 2 .
-Bulk density and 210 Pb xs profiles in sediment core BC4 from the outer Bay of Cádiz.