International Journal of Environmental Studies ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/genv20 Heavy metal mobility, bioavailability, and potential toxicity in sediments of the Korle lagoon in Ghana Richard Osae, Daniel Nukpezah, Daniel Amoako Darko & Adelina Mensah To cite this article: Richard Osae, Daniel Nukpezah, Daniel Amoako Darko & Adelina Mensah (2022): Heavy metal mobility, bioavailability, and potential toxicity in sediments of the Korle lagoon in Ghana, International Journal of Environmental Studies, DOI: 10.1080/00207233.2022.2042971 To link to this article: https://doi.org/10.1080/00207233.2022.2042971 Published online: 01 Mar 2022. Submit your article to this journal Article views: 76 View related articles View Crossmark data Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=genv20 INTERNATIONAL JOURNAL OF ENVIRONMENTAL STUDIES https://doi.org/10.1080/00207233.2022.2042971 Heavy metal mobility, bioavailability, and potential toxicity in sediments of the Korle lagoon in Ghana Richard Osae , Daniel Nukpezah, Daniel Amoako Darko and Adelina Mensah Institute for Environment and Sanitation Studies, University of Ghana, Accra, Ghana ABSTRACT KEYWORDS This study aimed to identify the mobility, bioavailability, and potential Heavy metals; mobility; toxicity of heavy metals (HM) in sediments of the Korle lagoon using toxicity; bioavailability; risk metal partition. Sediments were analysed using the BCR (Bureau of reference) method. As (60%), Pb (50%), Zn (57%) and Fe (40%) were found in high concentration in the ion exchangeable fraction. The oxidisable fraction was dominated by Cu (51%), Cd (95%), Cr (61%), Ni (63%), As (35%), Fe (21%), Hg (30%), Pb (19%) and Zn (11%). The reducible fraction was dominated by Zn (29%), Pb (24%) and Hg (19%). The residual fraction was dominated by Hg (49%) and Fe (34%). PCA was used to identify pH and organic matter as the most important factors controlling the mobility and bioavailability of HM in the sedi- ments. Arsenic, Pb, Zn and Fe were found to have the highest mobility. Chromium, Cu and Ni have medium mobility. Hg has low mobility. Most of the HM show potential bioavailability and toxicity. Introduction Pollution by HM in aquatic environments has become a major concern globally. This is largely caused by the increase in the use and processing of HM during various activities to meet the needs of the increasing population [1,2]. In aquatic environments, sediments serve as a reservoir for metals and play a major role in the distribution and storage of metals [3]. The Korle Lagoon stands out as one of the most polluted water bodies on earth [4]. The Korle Lagoon, however, used to be a freshwater ecosystem with an abundance of fishes, crabs, and other aquatic life forms that contributed to food security for the local residents [5]. The land uses within the fringes of the lagoon include settlements, the large concentration of industries such as textile factories and Ghana Breweries Ltd, vehicle repair shops, market, E-waste sites; and there is uncontrolled and unplanned urbanisation with many slums and squatters [6]. The Korle lagoon thus receives metals in untreated or inadequately treated wastewater from domestic, industrial, and agricultural sources. Previous studies recorded high concentrations of metals in and around the lagoon. Clottey [7] reported very high concentration for Cd (1.10–54.50 mg/kg), Pb (42.92–112.45 mg/kg), Cr (0.3–50.6 mg/ kg) and Cu (5.15–29.38) in the sediments of the Korle Lagoon. Aboagye [8] showed that the concentration of Cd and Pb in fishes obtained from the estuary of the Korle lagoon CONTACT Richard Osae osaerichard@ymail.com © 2022 Informa UK Limited, trading as Taylor & Francis Group Published online 01 Mar 2022 2 R. OSAE ET AL. exceeded the WHO recommended level and so was not safe for human consumption. Fosu-Mensah et al. [9] also reported that the concentration of Pb (84.44 mg/kg), Cd (103.66 mg/kg), Cu (202.99), Ni (72 mg/kg) and Sn (705.32 mg/kg) in lands and vegetation around the Korle lagoon exceeded the WHO/FAO levels for agricultural soils. Although there are many studies on the HM pollution of the Korle Lagoon, they have mainly focused on the total concentration of metals in the sediments. The total concentra- tion of metals in sediments can be used as an indicator for the assessment of sediment contamination. But, this cannot provide sufficient information for tohe assessment of environmental impact [10]. This is because the environmental behaviour of metals in soils and sediments depends on their chemical forms that influence their mobility, bioa- vailability, and potential toxicity [11–13]. Thus, there is a need for environmental impact and risk assessment studies on the different chemical forms of HM in the Korle Lagoon. Heavy metal speciation using sequential extraction procedures is necessary. The principle of sequential extraction is based on selective extraction of HM in different fractions using specific solvents [14]. This has been successfully applied to partition HM in sediments into different binding forms; ion exchangeable and carbonate bound (F1), Fe-Mn oxide bound (F2), organic matter/sulphide bound (F3), and residual (F4) [14,15]. The most reactive and bioavailable fraction is the exchangeable and carbonate bound fractions that are weak structures that are easily broken down by ion-exchange mechan- ism [16]. The oxidisable and reducible fractions are potential mobility fractions whose structure can be broken down as a result of a changes in pH, redox, and decomposition. The residual components comprise the non-mobile fraction that can only be disrupted by strong acids [16]. Thus, the sum of the metals associated with exchangeable and carbo- nate-bound, reducible, and the oxidisable fractions determine the mobile fraction (MF) and the most bioavailable metals [10]. High mobility of metals in the Korle lagoon may have an impact on major economic activities around it. These include fishing at the estuaries of the lagoon, farming, and residential activities. Fishes and food products can accumulate heavy metals introducing them into the food chain. But, accumulation of these metals in organic tissues is highly dependent on metal mobility since non-mobile metals (metals associated with F4) do not pose a major environmental risk. Since most studies have focused on the total metal concentration, it is not possible to ascertain if these metals pose a major risk. It is necessary to determine the chemical forms of the metals to determine their risk to the environment. In this study, the concentrations of heavy metals (As, Cu, Hg, Pb, Cd,, Zn, Fe, Ni) in surface sediments of the Korle lagoon were analysed to determine their different chemical forms in order to assess their mobility, bioavailability and potential toxicity. Methodology Study location The study focuses on the Korle lagoon and its catchment area. The Korle Lagoon is in Accra, the capital of Ghana and stretches to the west of the city centre within longitude 05°35ʹN and latitude 00°06ʹW. It has an estimated catchment of 400 km2 and covers a surface area of about 0.6 km2 [6]. The Korle Lagoon is fed by the Odaw River which INTERNATIONAL JOURNAL OF ENVIRONMENTAL STUDIES 3 flows through Accra with a catchment area of about 60% of Accra metropolitan area [6]. This catchment serves as the location for storm water runoff resulting in the transport of pollutants to various areas such as farms and residential areas. Sampling and chemical analysis The sampling location was divided into four quadrants and three surface sediments (1– 15 cm) were randomly collected from each quadrant (Figure 1) making a total of 12 samples. Sediments were sampled with a Van Veen grab sampler into plastic bags and transported to the laboratory in coolers containing ice packs. In the laboratory, samples were air-dried and ground into fine particles using a mortar and pestle and sieved through a 2 µm mesh and analysed for pH, Total organic carbon (TOC), and metals. Sequential extraction The sediment samples were analysed according to the BCR three-step sequential extrac- tion procedure described by Rauret et al. [17]. Fraction 4 was analysed according to the US EPA method 3050 B [18]. Exchangeable and carbonate-bound fraction (F1): 40 ml of 0.11 mol/ L acetic acid were added to 1 g of sample in a 50 mL centrifuge tube and extracted for 16 h at 25°C using a mechanical shaker followed by centrifugation at 3000 g for 20 min. The residue was washed with deionised (DI) water by shaking for 15 min, centrifuged and the supernatant discarded. Figure 1. Map of study area showing sampling location. 4 R. OSAE ET AL. Reducible fraction (Fe-Mn Oxide bound) (F2): 40 ml of 0.5 mol/L hydroxylamine hydrochloride, pH = 2, adjusted with 2.5% (v/v) nitric acid were added to the residue from the first step and extracted at room temperature using a mechanical shaker for 16 h. The extract was separated, and the residue was washed with deionised water as described in the first step. Oxidisable fraction (Organic matter and sulphide bound) (F3): 10 ml of Hydrogen peroxide were added in small aliquots to the residue from the second step. This was digested at room temperature for 1 h and further digested at 85°C for another hour. The volume was reduced to about 2 ml by heating the uncovered tube before adding another 40 ml of hydrogen peroxide. This was digested for 1 hr at 85°C to reduce the volume to about 1 ml. Samples were cooled to room temperature and 50 ml of 1 mol/L ammonium acetate, pH = 2, adjusted with nitric acid were added and extracted for 16 h. The residue was separated and washed as described in the previous step.8 Residual (F4): The residue from step 3 was then digested with Aqua Regia solution. One (1) ml of conc HCl and 1 ml of conc HNO3 was added to the residue and digested on a digestion block preheated to 110 ± 4°C for 40 minutes. After 40 mins, the tube was removed from the block, 10 ml of deionised water was added and digested for a further 20 mins. Samples were allowed to cool and topped up to the 50 ml mark. Heavy metal determination Nexion 2000- ICP-MS equipped with a micro mist nebuliser and microarray spray chamber was used to determine the concentration of metals. Calibration standards were prepared from a 500 mg/L multielement standard solution using a 1% (v/v) nitric acid solution. The instrument was calibrated from 1 to 100 µg/L using dwell times of 60, 25, 25,15,50,25,25,10,25 milliseconds for As, Cd, Cr, Cu, Hg, Ni, Pb, Fe and Zn respec- tively. Before analysis, samples were diluted at least 1:10 with 1% (v/v) HNO3 to reduce the high matrix. All the relative standard deviations of the replicate samples were <20%. Reagent blanks, duplicates, and certified reference material (SCP science Enviromat SS-2 soil standard) were used for quality control and accepted with recovery percen- tages >90%. pH Forty (40) ml of deionised water were added to 10 g of sample in a 50 ml tube. The mixture was placed on a mechanical shaker and mixed for 30 min at room temperature. After 30 min, samples were allowed to stand for 1 hour and pH was measured using a TPS Smartchem electrode probe and metre. Organic matter content An aliquot of dried sediments (0.35 g) was weighed into a carbon-free combustion boat and loaded onto an autosampler rack of LECO analyser and analysed for Total carbon. Another 0.35 g of dried sediments were weighed into a carbon-free combustion boat and treated with phosphoric acid until bubbling stops. The sample was dried in an oven pre-heated at 40°C for 24 hours and transferred to another oven preheated at 105°C. The boat was loaded onto the autosampler rack of a LECO analyser after drying and analysed for total inorganic carbon. Total organic carbon was calculated using equation 1. Organic matter was calcu- lated by multiplying TOC with the convenience/conversion factor of 1.724 [19]. INTERNATIONAL JOURNAL OF ENVIRONMENTAL STUDIES 5 TOCð%Þ ¼ Total carbon Inorganic carbon (1) Organic matterð%Þ ¼ TOCð%Þ � 1:724 (2) Risk assessment code (RAC) The Risk Assessment Code (RAC) was used to assess the contamination and potential toxicity of HM in the sediments of the Korle Lagoon. According to the RAC, HM in the sediments can thus be classified as no risk (RAC ≤ 1), low risk (LR) (1 < RAC ≤ 10), medium risk (MR) (10 < RAC ≤ 30), high risk (HR) (30 < RAC ≤ 50), and very high risk (VHR) (RAC > 50) [20]. The RAC was calculated as follows: RAC ¼ F1=TM� 100 (3) F1 – ion exchangeable fraction and carbonate bound fraction (mg/kg) TM – total concentration (mg/kg). Statistical analysis Exploratory data analysis to determine the mean, standard deviation, and percentages were done in Excel 2013. Results were expressed as mean ± Standard deviation. Principal Component Analysis (PCA) was used to reduce the dimensionality of data to make the data more interpretative and Correlation analysis was performed to determine the relationship between the variables. Both PCA and correlation were done using SPSS 24. Prior to statistical analysis, normality of data was checked using the Shapiro–Wilk test and data were found to have a normal distribution. Results Sediment properties and physicochemical characteristics Table 1, below, shows the basic physicochemical properties of the sediments. The sediments were alkaline with a pH range of 8.12–8.30. The concentration of organic matter (OM) in the bottom sediments from the Korle Lagoon was high and ranged from 6.5% to 7.6% (Table 1). A significant difference was observed in the concentrations from the different sampling sites for pH (P = 0.0021: P < 0.05) and organic matter (P = 0.037: P < 0.05) indicating that the pollutants were not evenly distributed in the sediments. Table 1. Mean organic matter and pH in surface sediments from the Korle lagoon. Site OM (%) pH 1 6.5 ± 0.06 8.13 ± 0.05 2 6.8 ± 0.01 8.30 ± 0.05 3 7.1 ± 0.41 8.30 ± 0.03 4 7.6 ± 0.03 8.21 ± 0.03 6 R. OSAE ET AL. Speciation of heavy metals Metal speciation was used to assess the heavy metal mobility and bioavailability in the Korle lagoon. Generally, heavy metals in the Korle lagoon were highly mobile as higher percentages were found in the mobile fractions; F1+ F2+ F3. Figure 2 shows the extractable percentages of heavy metals. Tables 2–10 illustrate the percentages of the metals in each fraction for the sediment samples. Arsenic was present in a higher percentage in the ion exchangeable fraction (60%). A little over one-third (35%) of As was found in the oxidisable fraction indicating that As is bound to organic matter and sulphides that are easily oxidised. The total mobile fraction of As (F1+ F2+ F3) was also high (96%) (Figure 2). But, As was hardly detected in the reducible fraction (0.3%) implying that there were low levels of As bound to Mn and Fe Oxides. Cadmium was present in higher concentrations in the oxidisable fraction F3 which is equivalent to 95% of the total Cd (10.15 mg/kg) concentration. This means that a high percentage of Cd is bound to organic matter and sulphides. The total mobile fraction for Cd was 96.11% indicating the Cd was highly mobile in the Korle Lagoon. It was hardly detected in F1 (0.9%), F2 (0.2%), and F4 (3%). Cu associated with F1, F2, and F3 constituted 36%, 5%, and 51% respectively. Cu was relatively high in F3 followed by F1 which indicates high mobility. The residual fraction represented 8% of the total Cu (18.75 mg/kg) content in the sediments. Ni was highly dominant in the oxidisable fraction (63%) followed by the ion- exchangeable and carbonate-bound fraction F1 (21%). F2 and F4 had extraction percen- tages of 5 and 11% respectively. The results show that the mobility of Ni was high. Similarly, Hg and Fe showed a high affinity for organic matter (F3 = 30% and 21% respectively). But, Fe was found to be dominantly associated with F1 (40%). Although Fe was high in the residual fraction F4, the total mobile fraction, F1+ F2+ F3 (66%) indicates that Fe is highly mobile in the Korle lagoon. Hg is highly associated with residual fraction F4 (49%) and may be less mobile compared to the other metals. It is, however, highly mobile in the Korle lagoon with the percentage mobile fraction of F1+ F2+ F3 being 50%. Pb and Zn are highly mobile elements in the Korle lagoon and are mostly associated with the ion-exchangeable and carbonate-bound fraction (50% and 57%, respectively). They are also highly associated with the reducible fraction (24 and 29%, respectively) and are bound to Fe and Mn oxides. They are present in high levels in the oxidisable fraction (19 and 11%) and are less associated with the residual fraction (5 and 4%). 120 100 80 60 40 20 0 As Cd Cu Cr Ni Fe Hg Pb Zn Exchangeable Reducible Oxidizable Residual Figure 2. Extraction percentages of As, Cd, Cu, Cr, Ni, Fe, Hg, Pb, and Zn in sampling sites in the Korle lagoon. Extracon percentage INTERNATIONAL JOURNAL OF ENVIRONMENTAL STUDIES 7 Table 2. Distribution of Arsenic in sediment fraction (mg/kg). Exchangeable F1 Acid reducible F2 Oxidisable F3 Residual F4 Total As Conc % Conc % Conc % Conc % F1+ F2+ F3 (%) 1 3.09 ± 0.86 1.33 ± 0.07 42.94 0.01 ± 0.04 0.36 1.69 ± 0.16 54.64 0.06 ± 0.02 2.05 97.95 2 1.98 ± 0.56 1.25 ± 0.04 63.11 0.02 ± 0.04 1.06 0.63 ± 0.07 32.09 0.07 ± 0.02 3.74 96.26 3 1.84 ± 0.59 1.29 ± 0.07 69.87 0.00 ± 0.01 0.00 0.48 ± 0.14 26.28 0.07 ± 0.02 3.85 96.15 4 2.16 ± 0.62 1.38 ± 0.07 64.09 0.00 ± 0.08 0.00 0.62 ± 0.15 28.78 0.15 ± 0.04 7.14 92.86 Mean 2.27 ± 0.56 1.31 ± 0.06 60.00 0.01 ± 0.01 0.35 0.86 ± 13.02 35.45 0.09 ± 0.04 4.20 95.80 Table 3. Distribution of Cd in sediment fraction (mg/kg). Exchangeable F1 Acid reducible F2 Oxidisable F3 Residual F4 Total Cd Conc % Conc % Conc % Conc % F1+ F2+ F3 (%) 1 10.35 ± 4.89 0.00 ± 0.01 0.01 0.02 ± 0.003 0.17 9.92 ± 0.21 95.86 0.41 ± 0.05 3.96 96.04 2 9.97 ± 4.65 0.10 ± 0.03 0.97 0.03 ± 0.01 0.28 9.47 ± 0.60 95.01 0.37 ± 0.04 3.73 96.27 3 9.99 ± 4.64 0.13 ± 0.04 1.26 0.02 ± 0.01 0.19 9.46 ± 0.23 94.67 0.39 ± 0.02 3.88 96.12 4 10.31 ± 4.77 0.13 ± 0.01 1.29 0.03 ± 0.01 0.31 9.73 ± 0.35 94.43 0.41 ± 0.02 3.97 96.03 Mean 10.15 ± 0.2 0.09 ± 0.06 0.88 0.02 ± 0.01 0.24 9.65 ± 0.22 94.99 0.39 ± 0.11 3.89 96.11 Table 4. Distribution of Cu in sediments fraction (mg/kg). Exchangeable F1 Acid reducible F2 Oxidisable F3 Residual F1+ F2 Total Cu Conc % Conc % Conc % Conc % + F3 (%) 1 9.09 ± 3.12 1.18 ± 0.022 12.93 0.09 ± 0.05 0.98 6.91 ± 0.55 75.98 0.92 ± 0.22 10.11 89.89 2 21.65 ± 4.51 9.51 ± 0.70 43.92 1.38 ± 0.12 6.36 9.13 ± 2.28 42.16 1.64 ± 0.35 7.56 92.44 3 20.21 ± 4.10 9.24 ± 0.68 45.69 1.29 ± 0.13 6.38 7.90 ± 1.66 39.08 1.79 ± 0.20 8.85 91.15 4 24.06 ± 5.28 10.06 ± 0.18 41.81 1.41 ± 0.03 5.86 11.08 ± 2.83 46.07 1.51 ± 0.05 6.26 93.74 Mean 18.75 ± 6.64 7.50 ± 4.23 36.09 1.04 ± 0.64 4.89 8.75 ± 1.8 50.82 1.46 ± 0.38 8.20 91.80 Table 5. Distribution of Cr in sediments fraction (mg/kg). Exchangeable F1 Acid reducible F2 Oxidisable F3 Residual F1+ F2 Total Cr Conc % Conc % Conc % Conc % + F3 (%) 1.00 41.48 ± 10.28 4.34 ± 0.11 10.45 4.34 ± 0.03 10.45 25.67 ± 2.03 61.89 7.14 ± 2.18 17.20 82.80 2.00 45.39 ± 11.41 5.79 ± 0.69 12.76 5.79 ± 0.24 12.76 28.46 ± 1.93 62.71 5.35 ± 0.50 11.78 88.22 3.00 45.68 ± 11.49 6.04 ± 0.42 13.22 6.02 ± 0.19 13.19 28.64 ± 1.52 62.70 4.98 ± 0.17 10.89 89.11 4.00 46.85 ± 11.13 6.80 ± 0.57 14.50 6.78 ± 0.06 14.48 28.36 ± 0.37 60.54 4.91 ± 0.44 10.48 89.52 Mean 44.85 ± 2.33 5.74 ± 1.03 12.73 5.73 ± 1.02 12.72 27.78 ± 1.41 61.96 5.59 ± 1.05 12.59 87.41 Table 6. Distribution of Ni in sediment fraction (mg/kg). Exchangeable F1 Acid reducible F2 Oxidisable F3 Residual F1+ F2+ F3 Total Ni Conc % Conc % Conc % Conc % (%) 1 11.55 ± 3.24 1.85 ± 0.08 16.01 0.54 ± 0.53 4.66 7.68 ± 0.24 66.45 1.49 ± 0.34 12.87 87.13 2 12.29 ± 3.00 2.96 ± 0.22 24.05 0.79 ± 0.23 6.46 7.35 ± 0.88 59.81 1.19 ± 0.22 9.68 90.32 3 11.14 ± 2.88 2.53 ± 0.08 22.74 0.47 ± 0.15 4.22 6.92 ± 0.24 62.09 1.22 ± 0.16 10.95 89.05 4 11.85 ± 3.16 2.62 ± 0.16 22.12 0.43 ± 0.07 3.65 7.51 ± 0.26 63.40 1.28 ± 0.13 10.84 89.16 Mean 11.71 ± 0.48 2.49 ± 0.46 21.23 0.56 ± 0.16 4.75 7.36 ± 0.33 62.94 1.30 ± 0.13 11.08 88.92 8 R. OSAE ET AL. Table 7. Distribution of Fe in sediment fraction (mg/kg). Exchangeable F1 Acid reducible F2 Oxidisable F3 Residual Total Fe Conc % Conc % Conc % Conc % F1+ F2+ F3 (%) 1 8476.52 ± 62 1965.12 ± 62 23.18 354.70 ± 29 4.18 4177.25 ± 504 49.28 1979.46 ± 159 23.35 76.65 2 11,067.35 ± 388 5098.57 ± 162 46.07 649.99 ± 138 5.87 1096.46 ± 659 9.91 4222.32 ± 163 38.15 61.85 3 11,032.52 ± 57 5188.54 ± 107 47.03 504.57 ± 76 4.57 778.63 ± 233 7.06 4560.78 ± 699 41.34 58.66 4 11,709.14 ± 101 5490.48 ± 101 46.89 584.78 ± 46 4.99 1900.32 ± 872 16.23 3733.56 ± 640 31.89 68.11 Mean 10,571.38 ± 1430 4435.68 ± 1655 40.79 523.51 ± 127 4.91 1988.17 ± 1533 20.62 3624.03 ± 1147 33.68 66.32 INTERNATIONAL JOURNAL OF ENVIRONMENTAL STUDIES 9 Table 8. Distribution of Hg in sediment fraction (mg/kg). Exchangeable F1 Acid reducible F2 Oxidisable F3 Residual F1+ F2+ F3 Total Hg Conc % Conc % Conc % Conc % (%) 1 0.21 ± 0.05 0.009 ± 0.002 4.37 0.03 ± 0.004 15.53 0.04 ± 0.02 18.93 0.13 ± 0.05 61.17 38.83 2 0.19 ± 0.04 0.00 ± 0.001 0.00 0.03 ± 0.005 16.15 0.08 ± 0.02 39.58 0.09 ± 0.01 44.27 55.73 3 0.19 ± 0.04 0.002 ± 0.001 0.00 0.02 ± 0.002 12.50 0.08 ± 0.01 40.10 0.09 ± 0.08 47.40 52.60 4 0.20 ± 0.04 0.003 ± 0.002 1.02 0.06 ± 0.029 32.49 0.04 ± 0.01 21.83 0.09 ± 0.08 44.67 55.33 Mean 0.20 ± 0.01 0.004 ± 0.002 1.35 0.04 ± 0.02 19.17 0.06 ± 0.02 30.11 0.10 ± 0.02 49.38 50.62 Table 9. Distribution of Pb in sediment fraction (mg/kg). Exchangeable F1 Acid reducible F2 Oxidisable F3 Residual F1+ F2 + F3 Total Pb Conc % Conc % Conc % Conc % (%) 1 16.83 ± 1.91 6.28 ± 0.28 37.31 3.91 ± 0.22 23.21 4.91 ± 0.21 29.16 1.74 ± 0.61 10.32 89.68 2 30.96 ± 6.03 15.51 ± 2.51 50.09 8.79 ± 0.33 28.41 5.43 ± 0.27 17.54 1.23 ± 0.04 3.96 96.04 3 34.37 ± 8.05 19.91 ± 3.11 57.94 8.07 ± 1.08 23.48 5.13 ± 0.31 14.94 1.25 ± 0.08 3.65 96.35 4 37.64 ± 8.24 21.78 ± 0.90 57.85 8.97 ± 0.45 23.83 5.62 ± 0.20 14.94 1.28 ± 0.08 3.39 96.61 Mean 29.95 ± 9.16 15.87 ± 6.91 50.80 7.44 ± 2.38 24.73 5.27 ± 0.32 19.14 1.37 ± 0.24 5.33 94.67 Table 10. Distribution of Zn in sediment fraction (mg/kg). Exchangeable F1 Acid reducible F2 Oxidisable F3 Residual F1+ F2+ F3 Total Zn Conc % Conc % Conc % Conc % (%) 1 118.87 ± 21 38.93 ± 0.42 32.75 55.59 ± 7.14 46.76 18.20 ± 2.64 15.31 6.16 ± 1.49 5.18 94.82 2 242.24 ± 57 102.26 ± 23.89 42.21 118.19 ± 96.22 48.79 16.35 ± 2.91 6.75 5.44 ± 6.15 2.25 97.75 3 142.66 ± 50 110.99 ± 16.95 77.80 12.22 ± 5.64 8.57 13.84 ± 1.52 9.70 5.61 ± 0.57 3.93 96.07 4 171.32 ± 57 128.99 ± 7.86 75.29 19.40 ± 8.92 11.32 17.34 ± 2.18 10.12 5.60 ± 0.36 3.27 96.73 Mean 168.77 ± 53 95.29 ± 39 57.01 51.35 ± 48 28.86 16.43 ± 1.89 10.47 5.70 ± 0.33 3.66 96.34 Correlation analysis To identify common characteristics, sources of metals, and to evaluate the potential of organic matter and pH to influence metal mobility, Pearson correlation analysis was performed. Strong linear correlation was found in As vs Hg (0.983), Fe vs Cu (0.997), Fe vs Cr (0.997), Fe vs Pb (0.986), Cr vs Pb (0.995), Cr vs Cu (0.989) and Pb vs Cu (0.995). In addition, As, Cd and Hg had a significant negative correlation (P < 0.05) with pH (−0.940 < r < −0.983). Similarly, a moderate correlation was observed between Cu, Cr, Fe, Pb, Zn, and pH (0.629 < r < 0.722) at a 0.05 significant level. Organic matter correlated positively with Cu, Cr, Fe, and Pb. 10 R. OSAE ET AL. Principal component analysis (PCA) Principal Component Analysis (PCA) was applied to the total metals and mobile fraction in the sediments. PC1 accounts for 72.7% of the total variance suggesting that PC1 is the important factor controlling the source and distribution of metals in the sediments. PC 2 accounts for 14.7% and PC 3, 12.6% of the total variance (Table 11). PC1 is highly loaded with Cu, Cr, Pb, Fe, As, MFCu, MFCr, MFPb, MFFe, HgMF, AsMF, and organic matter and PC2 is loaded with As, Cd, Hg, and pH (Table 12). Figure 3 shows the loading plot of the first three PCs. Here may be seen the characteristics and spatial distribution of the metals in space with the contribution of variables in the sample. The closeness of variables signifies the strength of their correlation. Table 11. Correlation between heavy metals (As, Cd, Cu, Cr, Ni, Fe, Hg, Pb, Zn) and pH and organic matter. As Cd Cu Cr Ni Fe Hg Pb Zn pH OM As 1 Cd .779 1 Cu −.889 −.455 1 Cr −.886 −.411 .989* 1 Ni −.081 −.073 .321 .190 1 Fe −.903 −.459 .997** .997** .246 1 Hg .983* .872 −.828 −.804 −.151 −.834 1 Pb −.890 −.409 .970* .995** .090 .986* −.800 1 Zn −.567 −.572 .630 .510 .843 .574 −.648 .431 1 pH −.940* −.945* .715* .686* .112 .722* −.984* .683* .629* 1 OM −.581 .056 .812* .871* −.015 .833* −.431 .886* .122 .269 1 *. Correlation is significant at the 0.05 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed). Table 12. Principal component analysis of heavy metals. Component OM 0.969 Pb 0.947 0.303 PbMF 0.944 0.311 Cr 0.934 CrMF 0.924 0.338 Fe 0.904 0.334 CuMF 0.888 0.354 Cu 0.884 0.321 0.34 HgMF 0.833 0.348 0.429 FeMF 0.761 −0.534 0.369 AsMF −0.748 −0.647 As −0.701 −0.701 Cd −0.981 pH 0.417 0.891 Hg −0.567 −0.797 Ni 0.997 NiMF 0.981 Zn 0.426 0.877 ZnMF 0.497 0.451 0.741 INTERNATIONAL JOURNAL OF ENVIRONMENTAL STUDIES 11 Figure 3. Loading plot of the first three principal components. Risk assessment The Risk Assessment Code (RAC) was used to evaluate the risk associated with the contamination of heavy metals in the sediments of the Korle lagoon. The RAC system considers the percentage of heavy metals present in F1 (the exchangeable and carbonate-bound forms). Figure 4 shows the risk analysis calculated with the concentration of metals in F1. Cd presented no risk in the sediments for all sample sites. Similarly, Hg presented no risk for all sampling sites except site 1 which was low risk (1 < RAC ≤ 10). Cu showed medium risk (10 < RAC ≤ 30) for sample site 1 and high risk (30 < RAC ≤ 50) for Site 2, 3, 4. Cr showed low risk for site 1 (RAC = 10) and medium risk (10 < RAC ≤ 30) for site 2, 3, 4. Similarly, Ni showed medium risk (10 < RAC ≤ 30). As shown, high risk (30 < RAC ≤ 50) for sampling site 1 and very high risk (RAC > 50) for Site 2, 3, and 4. Similarly, Zn showed high risk (30 < RAC ≤ 50) for sampling site 1 and 2 and very high risk (10 < RAC ≤ 30) for sampling site 3 and 4. As, Cu, Fe, Pb, and Zn fall under high risk to very high- risk category and can easily enter the food chain causing serious problems in the ecosystem. 12 R. OSAE ET AL. 90 Site 1 Site 2 Site 3 Site 4 80 Threshholf for Low risk Threshold for medium risk Threshold for high risk Threshold for very high risk 70 60 50 40 30 20 10 0 AS Cd Cu Cr Ni Fe Hg Pb Zn Metals Figure 4. Risk assessment code for metals in sediments of the Korle Lagoon. Discussion The results revealed that the HM in the sediments of Korle Lagoon are distributed between different geochemical fractions. The sequential extraction procedure adopted in this work allowed us to recognise that As, Cd, Cr, Pb, Ni, Cu, Fe, and Zn are significantly associated with forms such as carbonates, oxides of Fe and Mn and organic matter. Many studies have shown that organic matter plays a major role in the ability of bottom sediments to accumulate metals [10,21–23]. Sediments from the Korle Lagoon showed high concentrations of organic matter (6.5–7.6%). Because of the presence of negatively charged groups, organic matter has a high capacity to complex and adsorb to cations [10]. As a result, enrichment of trace metals in organic-rich sediments has been reported [21,24]. The rich organic matter in the study area can result in the formation of heavy metal-organic matter complexes reducing the bioavailability of metals [25,26]. The highest amount of metal for Cu (51%), Cd (95%), Cr (61%), Ni (63%) in the studied location was associated with F3. Similarly, As (35%), Fe (21%), Hg (30%), Pb (19%), and Zn (11%) also showed significant concentration in F3. Therefore, the mobility and bioavailability of HM in the Korle lagoon are highly influenced by organic matter. It has also been shown that metal-bound organic substances are highly influenced by the pH of the medium [27]. At low pH, trace metals are highly soluble and more bioavailable because of increased desorption and low adsorption. At medium pH, the pH-adsorption edge occurs resulting in increased adsorption of trace metals to organic substances [27]. At high pH, elements become completely adsorbed resulting in low bioavailability [27]. According to Kabata-Pandias [28], the solubility of most trace metals decreases with increasing pH. As such, pH values (8.1–8.3) will suggest that most of the metals tested in our work will be less available because of the alkaline pH of the sediments. Krol et al [29] have, however, shown that a significant amount of Pb and Cu can be released at alkaline pH (10.5). This agrees with our findings that showed that RAC INTERNATIONAL JOURNAL OF ENVIRONMENTAL STUDIES 13 even at high pH, some HM can have high mobility. High correlation was found between Cu and pH and Pb and pH (0.715 and 0.683 respectively). Similarly, Fe, Zn, and Cr also showed high correlation with pH (0.722, 0.629, 0.686 respectively). This suggests that the mobility of Pb, Cu, Fe, Zn, and Cr is highly influenced by pH. Arsenic (As), Cd, and Hg, however, had a negative correlation with pH (−0.940, −0.945, −0.980 respectively) which means that a substantial amount of Cd, As, and Hg would be released and become bioavailable following a reduction in the pH. Arsenic (As), Zn, Pb, and Fe have the highest possibility of being released from the sediment by a simple ion-exchange mechanism. Zn has the highest mobility since it presented the highest concentration in F1. Similar observations about Zn have been made by many authors [10,25]. High levels of Zinc in water occur as a by-product of steel production and from coal-fired power stations and burning of waste materials. The burning of electronic waste that occurs within the fringes of the Korle lagoon could be the source of Zinc in its waters. A significant correlation between Zn and pH has been reported [25,30] which agrees with our findings. Baran and Tarnawski [10],, however, did not observe any positive correlation between Zn and pH. Under reduced conditions, the release of Pb, Zn, and Hg is expected to be high because they were more associated with the reducible fraction F2. Principal component analysis (PCA) was applied to the present data obtained from the total HM and mobile fraction in sediments to identify the sources of pollution of the Korle lagoon. The first PC which accounts for 72.6% of the total variance is highly correlated with organic matter, total HM of Cu, Fe, Cr, Pb and the mobile fractions (CuMF, FeMF, CrMF, PbMF) of those elements with loadings of 0.905, 0.922,0.948, 0.956, 0.888, 0.904, 0.924, 0.944 respectively. PC1 also correlated with Hg (−0.567), AsMF (−0.748) and HgMF (0.833). The correlation of total HM, non-residual fraction, and organic matter indicates anthropogenic inputs [31,32]. Treated and untreated industrial sewage, municipal sewage, household waste, and waste from agricultural fields are the main sources of pollutants in the Korle Lagoon [4]. The different concentration of metals in the fractions indicates the different sources of pollution. To identify the sources and transport of heavy metals, Pearson’s correlation was carried out [33]. A significant positive correlation between Cu, Pb, Fe and Cr (0.683 < r < 0.997) implies that these elements have a common origin and are associated with increasing organic matter. The high correlation of Pb and Cu agrees with the findings of Zhang et al. [11], according to whom the sources of contamination are likely related to the input of pesticides and fertilisers. The negative or less correlation of the other metals indicates the different sources of contamination. Mostly, because of the weak association between the ion-exchangeable fraction and sediments, the high concentration of metals in this fraction is assumed to result from a more recent pollution. Similarly, those bound to reducible and oxidisable fraction indicate that pollution is less recent [3]. This means that, although the Korle Lagoon may have received recent pollution from As (F1 = 60%), Fe (F1 = 40%), Pb (F1 = 50%), and Zn (F1 = 57%), pollution from Cd, Cu, Cr, Ni, and Hg may be less recent as they are more associated with F2 and F3. Mitigation strategies to reduce pollution of the Korle Lagoon may have been unsuccessful as Korle still receives a significant amount of metals. This is 14 R. OSAE ET AL. corroborated by the significant concentrations of As, Cd, Cu, Cr, Ni, Pb, and Zn observed in F1 (recent) and F2+ F3 (less recent). There is continuous pollution from anthropogenic sources over time. The risk assessment code (RAC) was used to demonstrate that there was a very high risk of metal mobilisation from the sediments of the Korle lagoon. This implies that aquatic organisms and plants around the Korle Lagoon are at risk of HM accumulation. Lands and agricultural fields around the Korle lagoon has been shown to have high levels of heavy metals [9]. High concentration of metals in aquatic organisms obtained from the estuary of the Korle lagoon has also been reported [8]. This agrees with our finding that showed that metals stored in the sediments have high mobility and hence are highly available to be absorbed into the tissues of aquatic organisms and plants within the catchment area of the lagoon. Human exposure to heavy metals occurs through the consumption of contaminated water and food [34]. Consumption of food with a high concentration of heavy metals can cause adverse effects such as infertility, cardiovascular diseases, hormonal imbalance, neurological disorders, digestive tract disorders, and can- cer [35]. Conclusion The analysis of the concentrations of heavy metals and their chemical forms showed that sediments of the Korle lagoon are contaminated with heavy metals, from anthropogenic sources. This research revealed that all studied metals were in the highly mobile phases of the sediments. Organic matter and pH are the most important factors controlling heavy metal distribution and mobility in the sediment. Cu, Cr, Fe, Pb, and Cd were found to be dominant in sediments of the Korle Lagoon. Further, As, Zn, Pb, Cu, and Fe have the highest possibility of being released as they were more associated with the exchangeable and carbonate-bound fraction. Hg is less mobile, but mobility may increase following a reduction in pH. The risk analysis with the RAC indicated that As, Cu, Fe, Pb, and Zn fall under the high-risk to the very high-risk category and can easily enter the food chain. It is advisable to monitor the heavy metal concentration in aquatic biota and plants around the Korle Lagoon to assess the risk of consumption of such foods. The local residents should be educated about the dangers identified in this study. Disposal of waste should be conducted to appropriate standards, and not be haphazard. Sanitation should be according to modern hygienic requirements. Pollution control should be carried out according to law. There should be adequate treatment of waste before discharge. Remediation of the lagoon as well as the sediments and soils around the lagoon should take place in conjunction with these other measures. There should be epidemiological research into the health status and mortality of local residents in regard to the diseases found to be linked to exposure to heavy metals. Otherwise, the quality of the environ- ment and public health in the Korle Lagoon area cannot be guaranteed. Disclosure statement No potential conflict of interest was reported by the author(s). INTERNATIONAL JOURNAL OF ENVIRONMENTAL STUDIES 15 ORCID Richard Osae http://orcid.org/0000-0002-4570-4779 References [1] Masindi, V., and Muedi, K.L., 2018, Environmental contamination by heavy metals. In: Saleh, H.E.M, and Aglan, R.F. (Eds), Heavy Metals, Intech Open, pp115-133. 10.5772 intechopen. 76082. [2] Briffa, J., Sinagra, E., and Blundell, R., 2020, Heavy metal pollution in the environment and their toxicological effects on humans. Heliyon 6(9), 04691. doi:10.1016/j.heliyon.2020. e04691. [3] Abdallah, M.A.M. and Mohamed, A.A., 2019, Mobility and risk assessment of heavy metals by sequential extraction in coastal sediment south Mediterranean Sea, Egypt. Marine Systems and Ocean Technology 14, 42–50. doi:10.1007/s40868-019-00054-3. [4] Boadi, K. and Kuitunen, M., 2002, Urban waste pollution in Korle Lagoon. The Environmentalist 22, 301–309. doi:10.1023/A:1020706728569. [5] Squire, J., 2018, Mitigating the Korle Lagoon ecological pollution problem in Accra, Ghana, through a framework for urban management of the environment. In: L. Swatuk and C. Cash (Eds.) Water, Energy, Food and People across the Global South (Palgrave Macmillan, Cham: International Political Economy Series), pp. 117–134. doi:10.1007/978-3-319-64024-2_5. [6] Aglanu, L.M. and Appiah, D., 2014, The Korle lagoon in distress: The stress of urban solid waste on water bodies in Accra, Ghana. International Journal of Innovation and Applied Studies 7(2), 717–728. [7] Clottey, C.A., 2018. Assessment of the physichochemical parameters and heavy metal contamination in Korle and Kpeshie Lagoon. Master thesis. Institute for environment and sanitation studies, University of Ghana, Legon. [8] Aboagye, H.R., 2012, Heavy metal concentration in sediment and flesh of Boe Drum (Pteroscion peli) and greater Amberjack (Seriola Dumerelli) from the Korle Lagoon Estuary, Accra Ghana. Master thesis. College of Agricultural and natural resources, Department of fisheries and water shed management, Kwame Nkrumah University of Science and Technology. [9] Fosu-Mensah, B., Addae, E., Yirenya-Tawiah, D., and Nyame, F., 2017, Heavy metals concentration and distribution in soils and vegetation at Korle Lagoon area in Accra, Ghana. Cogent Environmental Science 3(1), doi:10.1080/23311843.2017.1405887. [10] Baran, A. and Tarnawski, M., 2015, Assessment of heavy metals mobility and toxicity in contaminated sediments by sequential extraction and a battery of bioassays. Ecotoxicology 24, 1279–1293. doi:10.1007/s10646-015-1499-4. [11] Zhang, Q., Han, G., Liu, M., Li, X., Wang, L., and Liang, B., 2019, Distribution and contamination assessment of soil heavy metals in the Jiulongjiang River Catchment, Southeast China. International Journal of Environmental Research and Public Health 16 (23), 4674. doi:10.3390/ijerph16234674. [12] Zhang, C., Yu, Z., Zeng, G., Jiang, M., Yang, Z., Cui, F., Zhu, M., Shen, L., and Hu, L., 2014, Effects of sediment geochemical properties on heavy metal bioavailability. Environment International 73, 270–281. doi:10.1016/j.envint.2014.08.010. [13] Wu, B., Wang, G., Wu, J., Fu, Q., and Liu, C., 2014, Sources of heavy metals in surface sediments and an ecological risk assessment from two adjacent plateau reservoirs. PLoS One 9(7), e102101. doi:10.1371/journal.pone.0102101. [14] Tessier, A.P. and Campbell, G.C., 1979, Kisson M. Sequential extraction procedure for speciation of particulate traces metals. Analytical Chemistry 51, 844–851. doi:10.1021/ ac50043a017. 16 R. OSAE ET AL. [15] Ure, A., Quevauviller, P., Muntau, H., and Griepink, B., 1993, Speciation of heavy metals in soils and sediments. An account of the improvement and harmonization of extraction techniques undertaken under auspices of the BCR of the Commission of the European Communities. International Journal of Environmental Analytical Chemistry 51, 135–151. doi:10.1080/03067319308027619. [16] Najamuddin, and Surahman, 2017, Dispersion, speciation, and pollution assessment of heavy metals Pb and Zn in surface sediment from disturbed ecosystem of jeneberang waters. IOP Conference Series: Earth and Environmental Science 89, 012030. doi:10.1088/ 1755-1315/89/1/012030. [17] Rauret, G., Lopez-Sanchez, J.F., Sahuquillo, A., Rubio, R., Davidson, C., Ure, A., and Quevauviller, P., 1999, Improvement of the BCR three-step sequential extraction procedure prior to the certification of new sediment and soil reference materials. Journal of Environmental Monitoring 1(1), 57–61. doi:10.1039/A807854H. [18] Environmental protection agency, 1996. Method 3050B: Acid Digestion of Sediments, Sludges, and Soils, Revision 2. [19] Walkley, A. and Black, I.A., 1934, An examination of the Degtjar method for determination of soil organic matter and a proposed modification of the chromic acid titration method. Soil Science 37(1), 29–38. doi:10.1097/00010694-193401000-00003. [20] Latosinska, J. and Czapik, P., 2020, The ecological risk assessment and chemical speciation of heavy metals in ash after the incineration of municipal sewage sludge. Sustainability 12 (16), 6517. doi:10.3390/su12166517. [21] Yang, X., Xiong, B., and Yang, M., 2010, Relationships among heavy metals and organic matter in sediment cores from Lake Nanhu, an Urban Lake in Wuhan, China. Journal of Freshwater Ecology 25(2), 243–249. doi:10.1080/02705060.2010.9665074. [22] Venkatramanan, S., Chung, S.Y., Ramkumar, T., and Selvam, S., 2015, Environmental monitoring and assessment of heavy metals in surface sediments at Coleroon River Estuary in Tamil Nadu, India. Environmental Monitoring and Assessment 187(8), 505. doi:10.1007/s10661-015-4709-. [23] Bhuyan, M.S., Bakar, M.A., Rashed-Un-Nabi, M., Senapathi, V., Chung, S.Y., and Islam, S. M., 2019, Monitoring and assessment of heavy metal contamination in surface water and sediment of the Old Brahmaputra river. Bangladesh. Applied Water Science 9, 125. doi:10.1007/s13201-019-1004-y. [24] Keene, A.F., Johnston, S.G., Bush, R.T., Burton, E.D., and Sullivan, L.A., 2010, Reactive trace element enrichment in a highly modified, tidally inundated acid sulfate soil wetland: East Trinity, Australia. Marine Pollution Bulletin 60(4), 620–626. doi:10.1016/j. marpolbul.2010.02.006. [25] Shaheen, S.M. and Rinklebe, J., 2014, Geochemical fractions of chromium, copper, and zinc and their vertical distribution in floodplain soil profiles along the Central Elbe. Geoderma 228–229, 152–159. doi:10.1016/j.geoderma.2013.10.012. [26] Li, G., Lu, N., Wei, Y., and Zhu, D., 2018, Relationship between heavy metal content in polluted soil and soil organic matter and pH in mining areas. IOP Conference Series: Materials Science and Engineering 394, 052081. doi:10.1088/1757-899x/394/5/052081. [27] Bradl, H.B., 2004, Adsorption of heavy metal ions on soils and soils constituents. Journal of Colloid and Interface Science 277(1), 1–18. doi:10.1016/j.jcis.2004.04.005. [28] Kabata-Pendias, A. and Pendias, H., 2011, Trace Elements in Soils and Plants, 4th, (Boca Raton: CRC Press). [29] Król, A., Mizerna, K., and Bożym, M., 2015, An assessment of pH-dependent release and mobility of heavy metals from metallurgical slag. Journal of Hazardous Materials 384, 121502. doi:10.1016/j.jhazmat.2019.121502. [30] Morillo, J., Usero, J., and Gracia, I., 2004, Heavy metal distribution in marine sediments from the southwest coast of Spain. Chemosphere 55, 431–442. doi:10.1016/j. chemosphere.2003.10.047. INTERNATIONAL JOURNAL OF ENVIRONMENTAL STUDIES 17 [31] Martínez-Santos, M., Probst, A., García-García, J., and Ruiz-Romera, E., 2015, Influence of anthropogenic inputs and a high-magnitude flood event on metal contamination pattern in surface bottom sediments from the Deba River urban catchment. Science of the Total Environment 514, 10–25. doi:10.1016/j.scitotenv.2015.01.078. [32] Sungur, A., Soylak, M., and Özcan, H., 2016, Chemical fractionation, mobility, and envir- onmental impacts of heavy metals in greenhouse soils from Çanakkale, Turkey. Environmental Earth Sciences 75, 334. doi:10.1007/s12665-016-5268-3. [33] Marina, P., Snežana, M., Maja, N., and Miroslava, M., 2020, Determination of heavy metal concentration and correlation analysis of turbidity: A case study of the zlot source (Bor, Serbia). Water Air Soil Pollution 231, 98. doi:10.1007/s11270-020-4453-x. [34] Shirani, M., Afzali, K.N., and Soleimani-Sardo, M., 2020, Pollution and contamination assessment of heavy metals in the sediments of Jazmurian Playa in southern Iran. Scientific Reports 10(1), 4775. doi:10.1038/s41598-020-61838-x. [35] Shirani, M., Akbari, A., and Hassani, M., 2015, Adsorption of Cadmium (ii) and Copper (ii) from soil and water samples onto a magnetic organozeolite modified with 2-(3,4-dihydroxyphenyl)-1,3-dithiane using artificial neural network and analysed by flame atomic absorption spectrometry. Analytical Methods 7, 6012–6020. doi:10.1039/ C5AY01269D.