The Prestea-Huni Valley Municipality served as the study location and it is one of the 14 MMDAs in the Western Region of Ghana. The municipality shares a northern border with the Wassa Amenfi East and Amenfi Central Districts. To the west of PHVMA is the Amenfi West District. The Municipality of Tarkwa Nsuaem is located in the southwest. The municipality’s administrative centre is in Bogoso. The PHVMA settlements included in this research were Bogoso, Prestea, Aboso, and Huni Valley. The map of the study area is shown by Figure 1. It is located within latitudes 5˚15'N - 5˚40'N and longitudes 1˚40'W - 2˚15'W. The municipality spans approximately 1200 km² ([17]). The area has a population of 229,301, of whom 117,744 are males and 111,557 are females ([5]).

Figure 1. Map of study area with sampling locations.

A total of nine representative soil samples were collected from four Final Waste Disposal Sites (FWDS) to evaluate heavy metal concentrations. A systematic judgmental sampling design was employed to ensure adequate spatial coverage across varying waste-age profiles and topographic features. The nine samples were distributed as follows: 2 samples from the Hunu Valley site, 3 from the Abosso site, 3 from the Prestea site, and one sample from the Bogoso site. Sampling points were selected based on proximity to active tipping faces and zones of leachate accumulation to represent a “worst-case” contamination scenario.

Points were spaced approximately 100 meters apart to capture spatial heterogeneity within the landfill boundaries. Sampling was conducted during the period. At each location, soil cores were extracted from a depth of 0 - 20 cm using a stainless-steel hand auger, with precise coordinates recorded through handheld GPS. To ensure sample integrity, non-soil debris—including metals, plastics, glass, and stones—was manually removed in the field. Samples were then homogenised, sealed in airtight polyethylene bags, and transported to the Environmental Monitoring Laboratory at the University of Mines and Technology (UMaT) for heavy metal analysis.

Descriptive statistics, including arithmetic means and standard deviations, were computed for each landfill site to characterise heavy metal distribution. For the Bogoso site, where the sample size (n = 1) precluded statistical dispersion analysis, reported values were treated as point-source concentrations. Given the high spatial proximity of sampling points, mean values were utilised for spatial distribution mapping. To assess the variability of heavy metal concentrations, the Coefficient of Variation (CV%) was calculated and categorised as moderate (CV ≤ 50%), high (50% < CV ≤ 100%), or extraordinarily intense (CV > 100%) according to ([25]).

Handling of Data below Detection Limits

To ensure statistical integrity for the calculation of means, CV%, and subsequent pollution indices, data points recorded below the method detection limit (LOD) were handled using the full detection limit (LOD), providing a conservative estimate of heavy metal loading.

Heavy metal concentrations in the soil samples were determined following standard acid digestion protocols. Dried and pulverised soil samples (1.0 g each) were subjected to aqua regia digestion using a 1:3 mixture of concentrated HNO₃ and HCl. The samples were digested at 105˚C in conical flasks until the volumes were reduced. After cooling to room temperature, the digests were quantitatively transferred to 100 mL volumetric flasks, diluted to the mark with deionised water, and filtered through Whatman No. 42 filter paper. The resulting filtrates were analysed for Arsenic (As), Zinc (Zn), Cadmium (Cd), and Lead (Pb) using an Atomic Absorption Spectrophotometer (AAS). The instrument was calibrated using analytical-grade multi-element standard solutions to establish linearity, with a method detection limit of 0.002 mg/kg.

Quality Assurance and Quality Control (QA/QC)

To ensure the accuracy and reproducibility of the analytical data, rigorous QA/QC protocols were implemented throughout the digestion and determination phases. Procedural blanks were prepared alongside the batch samples to monitor for reagent-grade contamination and potential carry-over effects during the analytical process. Precision was assessed through replicate analyses, with relative percent difference (RPD) values maintained below 5%. Analytical accuracy was verified using certified reference materials (CRMs) to determine the percentage recovery of each heavy metal; recovery rates were maintained within the acceptable range of 95% - 110%. Instrument stability was further verified by periodically re-analysing mid-range calibration standards every ten samples to check for potential baseline drift.

Geostatistical methods were employed to visualise the spatial distribution of soil contamination ([8]). Specifically, Inverse Distance Weighting (IDW) within ArcGIS (version 10.3) was utilised to generate maps showing heavy metal concentrations at various sampling points across the study’s towns. Additionally, IDW provided a visual representation of the Possible Ecological Risk Index (PERI) results, which clarified the distribution of heavy metals in all target locations and highlighted the potential ecological threats that the landfills posed to the surrounding communities. Also, the sources of soil pollution causing heavy metal contamination and the relative contributions of each source were revealed by geographical analysis ([9]).

The Pollution Load Index (PLI) ([22]), the Geoaccumulation Index (I_geo) ([33]), the Nemerow pollution index (NPI) ([21]), and the Potential Ecological Risk Index (PERI) ([20]) were used to evaluate the heavy metal contamination in landfill soils at PHVDA. The I_geo is mainly used to evaluate contamination of certain heavy metals in relation to the concentration of the geochemical background, whereas PLI is used to determine the pollution level for all heavy metals. I_geo is mainly used to evaluate contamination of certain heavy metals in relation to the concentration of the geochemical background, whereas PLI is used to determine the pollution level for all heavy metals. The soil quality was thoroughly assessed using NPI. The environmental hazards posed by the heavy metals were assessed using the PERI. The following formulas were used to determine these indicators:

PLI = (Pl₁ × Pl₂ × Pl₃ × Pl₄ × … × Pl_n)^1/n (3)

The pollution index (PI) for heavy metals was calculated by comparing the concentration of each heavy metal in the soil (mg/kg) with its corresponding background value (also mg/kg). These background values for soil were obtained from the Canadian Soil Quality Guidelines for the Protection of Environment and Human Health (2001); n represents the number of heavy metals, and b represents the biological toxicity factor for each heavy metal ([12]). The Canadian Soil Quality Guidelines were utilised due to the absence of specific, comprehensive national soil quality standards for heavy metals in Ghana, thereby providing a robust and internationally recognised benchmark for environmental risk assessment in this study. The specific background values used for each heavy metal were: Arsenic (As): 12.0 mg/kg; Zinc (Zn): 200.0 mg/kg; Cadmium (Cd): 1.4 mg/kg; and Lead (Pb): 70.0 mg/kg. The biological toxicity factors (T_r) used specifically for the Potential Ecological Risk Index (PERI) were adapted as follows: Arsenic (As): 1; Zinc (Zn): 1; Cadmium (Cd): 30 and Lead (Pb): 5 ([12]). The assessment methods used are shown in Table 1.

Table 1. Assessment methods of PLI, I_geo, PERI and NPI.

The degree of association between heavy metals was evaluated using the Pearson correlation coefficient (PCC) ([13]). There may be a significant correlation between heavy metals with P < 0.05 (two-tailed). The interrelationships among heavy metals suggest they come from comparable geographic regions or share similar geochemical features ([34]). Additionally, Hierarchical Cluster Analysis (HCA) was carried out utilising the standardised data, using Z-scores and Ward’s method with squared Euclidean distances. HCA and PCC analyses were performed in OriginPro 2023 to identify likely sources of soil metals in the landfill soil.

As shown in Table 2, the mean concentrations of Zn at both the Aboso (2236.51 mg/kg) and Prestea (1192.04 mg/kg) landfills exceeded the Canadian Soil Quality Guidelines’ permissible agricultural/residential limit of 200 mg/kg (and the commercial/industrial threshold of 360 mg/kg). Similarly, the mean Pb concentration at the Aboso landfill (1548.73 mg/kg) significantly exceeded the permissible environmental limit of 140 mg/kg ([6]). Conversely, the mean concentrations of As and Cd remained below their respective environmental limits across all sites, though localised hotspots still present potential exposure risks ([7]). Lead is among the least mobile metallic elements in soil, making its accumulation highly persistent and hazardous ([7]). Its elevated concentration is often associated with the disposal of household dry-cell batteries and electronic waste in municipal waste streams ([9]). The high concentrations of Zn and Pb in the Aboso and Prestea soils are likely driven by localised small-scale mining discharges, discarded galvanised building materials, and the indiscriminate disposal of consumer electronics, batteries, and lead-based paints/coatings ([7]; [9]).

Table 2. Heavy metal concentrations in Landfill Soil (mg/kg).

Concentrations of Zn in soil samples at Aboso (2236.510 ± 3617.577 mg/kg) and Prestea (1192.04 ± 830.654 mg/kg) were higher than the permissible limit (1000 mg/kg) of the Canadian Soil Quality Guidelines for the Protection of Environment and Human Health. Thus, the high Zn concentration in the Aboso and Prestea soil samples may be attributed to mining activities and household waste (zinc-containing products such as cosmetics, coatings, and paints) ([32]). Also, when improperly disposed of, building materials, such as galvanised metal with a zinc coating to prevent corrosion, can transfer Zn into the landfill’s soil ([27]). According to ([23]), Zn pollution could also be attributed to the indiscriminate disposal of pesticides and fertilisers. In this study, Aboso P2 recorded the highest Pb concentration (4354.447 mg/kg), followed by Aboso P1 (283.044 mg/kg). Pb concentration at Aboso P1 was 8.709 mg/kg. The average lead concentration in Aboso’s soil samples was 1548.733 ±243.688 mg/kg, was well above the 200 mg/kg guideline established by the Canadian Soil Quality Guidelines for the Protection of Environment and Human Health (2001). This was lower than the mean Pb concentration (4400 mg/kg) recorded at a landfill site in Tangier (Morocco) in a study by ([7]).

Figure 2 illustrates the spatial distribution of the four target heavy metals (As, Zn, Cd, and Pb) across the sampling locations, utilising an Inverse Distance Weighting (IDW) interpolation method where a colour gradient from red to blue signifies high to low concentration zones, respectively. Despite the presence of visible localised hotspots, statistical analysis reveals no significant differences (p > 0.05) in overall metal distribution between the sites.

The mean concentrations of As decreased in the order of Prestea (5.39 mg/kg) > Bogoso (5.05 mg/kg) > Aboso (2.99 mg/kg) > Huni Valley (0.87 mg/kg). Elevated As levels at Prestea and Bogoso are primarily attributed to the local geology, characterised by auriferous and manganese-rich formations of the Birimian Supergroup that naturally release arsenic during weathering ([32]). This geogenic baseline is further augmented by anthropogenic inputs from industrial, household, and agricultural waste, specifically arsenic-based pesticides.

For Zn, the mean concentrations followed the decreasing order of Aboso (2236.51 mg/kg) > Prestea (1192.04 mg/kg) > Bogoso (617.29 mg/kg) > Huni Valley (26.43 mg/kg). The elevated Zn levels at Aboso and Prestea reflect active mining operations and associated mechanical waste; furthermore, the regional lithology is known to possess naturally high zinc concentrations ([18]).

The mean concentrations of Cd decreased in the order of Bogoso (2.12 mg/kg) > Prestea (1.46 mg/kg) > Aboso (0.34 mg/kg) > Huni Valley (<0.002 mg/kg). These elevated levels at Bogoso and Prestea are strongly linked to the improper disposal of Ni-Cd batteries, plastics, and glass ([7]).

Finally, the mean Pb concentrations decreased in the order of Aboso (1548.73 mg/kg) > Bogoso (67.49 mg/kg) > Prestea (32.66 mg/kg) > Huni Valley (<0.005 mg/kg). The exceptionally high Pb concentrations at Aboso are largely driven by the high density of local auto-mechanic and vulcanising workshops that utilise and subsequently dispose of lead-acid car batteries at the landfill site ([9]). Figure 2 is the Maps showing the spatial distribution.

Figure 2. Maps showing the spatial distribution of: (A) Zn, (B) As, (C) Cd, (D) Pb.

Soil Contamination Risk

Soil pollution indices, including the Geo-accumulation Index (I_geo), Pollution Load Index (PLI), and Nemerow Pollution Index (NPI), were evaluated to quantify overall contamination levels (Figure 3). I_geo values showed no significant variation (p > 0.05) in heavy metal contamination across the four landfills. For As, Cd, and Zn, I_geo values fell into the “unpolluted” class across all sites. However, Pb displayed an I_geo value of 2.37 at the Aboso site, indicating moderately to strongly polluted conditions, while remaining in the unpolluted class at Prestea (-3.19), Huni Valley (−15.87), and Bogoso (−2.15). This localised lead pollution is directly tied to the improper disposal of lead-acid batteries from local mechanical and vulcanising workshops ([9]).

This spatial variation is supported by the PLI results, which exceeded the critical pollution threshold of 1.0 only at the Aboso landfill (PLI = 2.54), indicating a highly polluted state. In contrast, the other three landfills fell well below the threshold, indicating low pollution loading: Prestea (0.53), Bogoso (0.51), and Huni Valley (0.01). The exceptionally low PLI at Huni Valley is attributed to the relatively young age of that specific landfill facility ([1]).

The Nemerow Pollution Index (NPI) confirmed these findings. Based on the assessment criteria, Aboso (2.4) and Prestea (2.2) were classified as moderately to heavily polluted. Under the NPI classification, Bogoso (1.3) was categorised as slightly polluted, while Huni Valley (1.0) reached the precautionary warning limit, indicating the need for immediate management interventions to prevent further degradation. The high pollution status at Aboso and Prestea is likely linked to human activities and the fact that these landfill sites have been operational for over two decades ([1]). At Aboso, Pb (1548.73 ±243.688 mg/kg) and Zn (2236.15 ±3617.577 mg/kg) were the primary contributors to the pollution status due to their high concentrations. In contrast, in Prestea, Zn (1192.04 ±830.654 mg/kg) was the principal pollutant.

Figure 3. (A) PLI of heavy metals, (B) PERI of heavy metals, (C) Heat map of I_geo levels, and (D) NPI of heavy metals across the various locations.

The Potential Ecological Risk Index (PERI) was calculated to synthesise the toxicity of individual heavy metals into a comprehensive risk profile, as illustrated in Figure 3. At the site level, the total PERI values decreased in the following order: Aboso (10020.21, representing an extremely high ecological risk) > Prestea (1452.83, indicating high risk) > Bogoso (1068.86, indicating high risk) > Huni Valley (35.21, indicating low risk).

Detailed risk indices for individual heavy metals are also shown in Figure 3, with the highest contributions observed as: Pb (7743.67) > Zn (2236.51) > As (29.83) > Cd (10.21) at Aboso; Zn (617.29) > Pb (337.46) > Cd (63.60) > As (50.51) at Bogoso; Zn (26.42) > As (8.70) > Cd (0.06) > Pb (0.025) at Huni Valley; and Zn (1192.04) > Pb (163.29) > As (53.85) > Cd (43.65) at Prestea.

These findings confirm that the Aboso landfill presents the most severe ecological threat to surrounding soils, vegetation, and groundwater, primarily driven by its exceptionally high Pb (T_r = 5) and Zn loading. This elevated risk correlates strongly with the landfill’s older operational age and the high volume of municipal and industrial waste it receives. The comparative spatial indices for PLI, PERI, and NPI are graphically presented in Figure 3.

The resulting correlation matrix (Figure 4) indicated a strong positive correlation between Arsenic (As) and Cadmium (Cd) (r = 0.68). This suggests that As and Cd likely share homologous characteristics and may originate from the same anthropogenic source. The HCA results further confirmed this. The dendrogram delineated two distinct clusters that exhibited dissimilar geochemical associations, as depicted in Figure 4. This suggests that heavy metals from comparable input sources belong to the same cluster ([9]). The squared Euclidean distance was used as the similarity measure; accordingly, heavy metals that appear close together in the resulting dendrogram are considered to share closely associated input sources.

In contrast, long-distance depicts dissimilarities ([3]). Figure 4 suggests that the toxic metals Arsenic (As) and Cadmium (Cd) grouped in cluster 1, likely represent a primary source of contamination. The findings reveal that improper waste disposal, including fertilisers and pesticides, could enrich the soil with As and Cd ([9]). Zn and Pb were loaded in cluster 2, with a long squared Euclidean distance, as shown in Figure 4, indicating dissimilar sources ([3]). Figure 4 shows the results of two different statistical analyses. The Pearson Correlation Coefficient is a measure of the correlation between two variables. The hierarchical cluster analysis is a method for grouping similar variables. Multivariate Statistical Analysis of Heavy Metals in Landfill Soil is shown in Figure 4.

Figure 4. Multivariate statistical analysis of heavy metals in landfill soil: Pearson correlation matrix and hierarchical cluster analysis (HCA) dendrogram.