Accumulation and distribution of mercury in cacao (Theobroma cacao) grown in artisanal and small-scale gold mining areas, Ashanti Region, Ghana

ABSTRACT

Mercury is used in artisanal and small-scale gold mining (ASGM) to extract gold by forming an amalgam. This raises considerable concern about mercury pollution in the surrounding environment and plants. This study aimed to investigate mercury concentrations in soil and cacao in the ASGM area of Ghana and to evaluate the human health risk of mercury in cacao. Agricultural soil and cacao (root, leaf, bean, and pod) samples were collected from three ASGM communities (Manso Abore, Nweneso No. 1, and Nweneso No. 2) in the Ashanti Region of Ghana. The mean total mercury concentrations in the soil were 76±36 µg/kg dry weight (n=42), corresponding to moderate contamination by the contamination factor. In the cacao samples, the highest mean total mercury concentration was found in the roots (211 µg/kg dry weight, n=12) and lowest in the beans (27 µg/kg dry weight, n=12). The bioaccumulation and transfer factors indicated a high accumulation of mercury in the roots of cacao (bioaccumulation factor=3.1) from the soil but a lesser accumulation in the aerial parts from the roots (transfer factor<1). The health risk assessment from the ingestion of cacao beans indicated no possibility of noncarcinogenic health effects (hazard quotient<1). Continuous mercury pollution may lead to accumulation along the food chain, which may be detrimental to human health. Therefore, strict measures and monitoring in ASGM communities are needed to protect human health.

INTRODUCTION

Mercury (Hg) is toxic to human health and is one of the top ten chemicals of most concern in public health (World Health Organization, 2020). Of the anthropogenic sources of Hg in the world, artisanal and small-scale gold mining (ASGM) is the largest emitter of Hg to the atmosphere, accounting for approximately 38% of the total Hg emissions (United Nations Environment Programme [UNEP], 2019). In the ASGM areas, Hg is used in an amalgam with gold to extract gold from ore and is released through the open burning of the amalgam and the use of cyanide with mercury-containing tailings.

In Ghana, gold mining has a long history (Gyamfi et al., 2022). According to the Minamata Convention on the Initial Assessment Report for Mercury in Ghana (Environmental Protection Agency [EPA], Ghana, 2018), the total input of Hg in Ghana was estimated to be 81.1 tons/year, of which mercury-amalgamated gold mining accounted for approximately 45.2 tons/year. However, gold extraction with mercury amalgamation accounted for 50% of the total Hg release in Ghana, and the amount of Hg released into the air, water, and land was estimated to be 32.5, 6.5, and 6.1 tons/year, respectively (EPA, Ghana, 2018). Yevugah et al. investigated the Hg concentration in soil samples collected systematically across the entire country of Ghana. They reported that high regional concentrations had been significantly increased by mining pollution (Yevugah et al., 2021; 2023). High concentrations of Hg in soil around the ASGM areas, including the Wassa West District, Western Region, Upper East Region, and Ashanti Region, have been reported (Tetteh et al., 2010; Rajaee et al., 2015; Mantey et al., 2020; Addai-Arhin et al., 2023).

High Hg concentrations in the soil in ASGM communities may lead to an accumulation in vegetables and crops grown in that soil, particularly plantain and cassava (Bortey-Sam et al., 2015; Addai-Arhin et al., 2023; Kithure and Hudumu, 2023). Ghana is the second leading producer of cacao and contributes approximately 20% of the total cacao production in the world (Ahoa et al., 2021). Notably, the cacao industry contributes about 3% of Ghana’s gross domestic product; 30% of the livelihood of the entire population depends on it (Ahoa et al., 2021). Heavy metals such as cadmium (Cd) and lead (Pb) have frequently been measured in cacao beans and their products because the Codex Alimentarius has established maximum acceptable levels for these metals (Mounicou et al., 2003; Amankwaah et al., 2015; Arham et al., 2017; Anyimah-Ackah et al., 2021; Barraza et al., 2021). However, existing reports on the Hg concentration in cacao beans are fewer than those on other metals such as Cd and Pb. Several recent studies have focused on elevated total Hg (THg) levels in cacao beans (Anyimah-Ackah et al., 2021), particularly in areas with ASGM (Osman et al., 2022). Therefore, further study is required to evaluate the transfer of Hg from agricultural soil to cacao around ASGM communities.

The present study aimed to evaluate the transfer of THg from agricultural soil to cacao and the potential human health risk associated with the ingestion of cacao products in ASGM communities in the Ashanti Region of Ghana. We investigated the THg concentration and bioavailable Hg concentration in agricultural soil and cacao samples and analyzed the relationship between the THg concentration in the soil and cacao root, leaf, pod, and bean samples.

MATERIALS AND METHODS

STUDY AREA AND SAMPLING METHODS

Sampling was conducted in three ASGM communities in the Ashanti Region of Ghana (Fig. 1)—Manso Abore in Amansie West District and two areas in Nweneso in Atwima Kwanwoma District—because these districts have a long history of cacao farming and ASGM. Gold mining activity with the use of mercury is in operation in these areas.

Fig. 1 Sampling sites in three different ASGM communities in the Ashanti Region

In March (the dry season) of 2023, 28 agricultural soil samples were randomly collected from different farms in Manso Abore (n=8) in Amansie West District and from Nweneso No. 1 (n=10) and Nweneso No. 2 (n=11) in the Atwima Kwanwoma District (Fig. 1) using a stainless steel cutlass at a depth of 0–20 cm. In March 2024, soil and cacao roots, leaves, pods, and beans were randomly sampled from 12 different cacao farms from the same three ASGM communities [Manso Abore (n=4), Nweneso No.1 (n=5) and No.2 (n=3)] as in 2023 (Fig. 1). Additionally, as the reference sample, soil and cacao samples were collected from three different farms in the Jamasi area (at least 56 km away from Manso Abore, Nweneso No. 1, and Nweneso No. 2), and composite samples of soil (n=1) and cacao (n=1) were used for THg analysis. This area was selected because of its low level of anthropogenic activity and lack of ASGM. Mercury contamination is considered to be negligible in this area. Permission to collect samples was obtained from the farm owners and community leaders.

SAMPLE TREATMENTS

Overall, 94 samples from all sites (42 soil samples and 52 cacao samples) were analyzed. Soil samples were air-dried for 72 hours, passed through a 2 mm sieve, and homogenized. The cacao beans were fermented for 4 days according to the farmers’ practice by keeping the fresh beans in sealed zip-lock bags before drying. All cacao crop samples except the beans were washed twice with tap water, then three times with deionized water, and subsequently sun-dried for 2 weeks. The samples were ground with a coffee grinder, passed through a 150 µm sieve, homogenized, and stored at 4°C until analysis. The grinder was washed with detergent, distilled water, and propanol after every use to prevent cross-contamination.

ANALYSIS OF HG CONTENT IN SOIL AND CACAO CROP SAMPLES

Approximately 20 mg of soil and cacao crop samples were measured in triplicate using a direct mercury analyzer (MA-3000, Nippon Instrument Corporation, Tokyo, Japan) to obtain the THg. The THg concentrations in the agricultural soil and cacao crop samples were expressed as µg/kg dry weight (µg/kg dw). Soil organic matter (SOM) was measured by the loss-on-ignition method in an electric furnace at 550°C, and the moisture content in all samples was determined after drying the soil samples and cacao crop samples in an oven for 12 h at 105°C and 45°C, respectively. The soil pH was measured using a calibrated pH meter (AS 800, AS ONE Corporation, Osaka, Japan) after mixing a 1:2.5 soil–deionized water suspension.

For the determination of the bioavailability of Hg in the soil where the cacao crop samples were collected, 1.00 g of soil was weighed into a 15 mL plastic centrifuge tube, and 0.1 mol/L of HCl with 1% CuSO₄ solution was added (Liu et al., 1997). The resulting mixture was shaken for 2 h and centrifuged for 30 min at 25°C at a speed of 3,800 rpm. The bioavailable Hg concentration in the supernatant was then determined using a direct mercury analyzer.

QUALITY CONTROL AND QUALITY ASSURANCE

A blank analysis was performed in triplicate for each batch of 10 samples. The THg concentrations of all samples were higher than the detection limit (0.3 µg/kg). For the accuracy and validation of the method analysis, certified reference material (NMIJ 7302-a, National Metrology Institute, Japan) for marine sediments and ERM-CC580 (Institute of Reference Materials and Measurements, Belgium) for estuarine sediments were used. The THg recovery rates for NMIJ 7302-a and ERM-CC580 ranged from 87% to 105% (n=5) and from 98% to 106% (n=3), respectively. The additive recovery rate using the mercury standard solution (100 pg/μL, 200 μL) for the cacao samples was 95%.

STATISTICAL ANALYSIS

Microsoft Excel 2013 (Microsoft Corporation, USA), OriginPro 2023b (OriginLab Corporation, USA), and IBM SPSS Statistics version 26 (IBM Corporation, New York, USA) were used for the statistical analysis. Shapiro–Wilk and Kolmogorov–Smirnov normality tests were used to determine the distribution of THg concentration. The Friedman analysis of variance (ANOVA) was used to determine the significant difference for the cacao crop samples (P<0.05). Spearman’s rho correlation between THg and bioavailable Hg in soil and cacao crop samples was performed. A principal component analysis (PCA) was conducted using the open-source statistical software R version 4.5.0 (R Core Team, 2025). The concentrations of THg and bioavailable Hg, pH of the soil, SOM, and THg concentrations in four parts (bean, root, leaf, and pod) of the cacao plant for 12 sites were used for the analysis. PCA was performed using the prompt function with a correlation matrix.

THE CONTAMINATION FACTOR

The contamination factor (CF) was employed to assess the THg contamination levels in the soil relative to the uncontaminated soil at the reference site using the equation below.

  

$$CF=C_S/C_{ref}$$(1)

where C_s is the THg concentration in the agricultural soil in the study area (µg/kg dw), and C_ref is the THg concentration in the reference soil (µg/kg dw). The CF values were classified as low contamination (CF<1), moderate contamination (1≤CF<3), considerable contamination (3≤CF<6), and very high contamination (CF≥6) (Gyamfi et al., 2022).

POTENTIAL ECOLOGICAL RISK INDEX (PERI)

The PERI was employed to evaluate the degree of ecological risks of the soil by considering the THg toxicity as proposed by Hakanson (1980). It was calculated from the CF using the toxicological factor for THg (T_r=40), as in equation 2.

  

$$E_r=T_r\times CF$$(2)

The risk factors were classified as low potential ecological risk (E_r<40), moderate potential ecological risk (40≤E_r<80), considerable potential ecological risk (80≤E_r<160), high potential ecological risk (160≤E_r<320), and very high ecological risk (E_r≥320).

BIOACCUMULATION AND TRANSFER FACTORS

The bioaccumulation factor (BAF) is an indicator that estimates the level of THg that moves from the soil to the cacao plant root. It indicates the extent of metal enrichment of the cacao plant from the soil, whereas the transfer factor (TF) assesses the ability of the cacao to transfer the THg from the root to the areal parts of the cacao crop (Yoon et al., 2006). The BAF and TF of THg were defined using the following equations.

  

$$BAF=C_{root}/C_{soil}$$(3)

  

$$TF=C_{aerial}/C_{root}$$(4)

where C_root is the THg concentration in the root (µg/kg dw), C_soil is the THg content in the agricultural soil, and C_aerial is the THg concentration in the aerial components (leaves, pods, and beans) (µg/kg dw). BAF or TF>1 indicates high transfer and absorption levels, whereas BAF or TF<1 indicates low transfer and absorption levels.

HEALTH RISK ASSESSMENT OF CACAO BEANS

In this study, ingestion was assumed to be the only route to cacao exposure to THg. Equation 5, as used by Olawale et al. (2023), was used to determine the estimated daily intake (EDI) of THg through the consumption of cacao beans and related cacao products.

  

$$\text{EDI}=C_{beans}\times DI/BW$$(5)

where C_beans represents the THg concentration in cacao beans (μg/kg dw), and DI represents the daily intake of cacao (kg/day). The cacao consumption per capita in Ghana is 0.52 kg/year (GCB Strategy & Research Dept, 2023); a body weight (BW) of 70 kg was used. For the health risk assessment, the worst scenario was assumed, i.e., 100% maximum concentration transfer from cacao beans to the finished product.

The hazard quotient (HQ) was estimated by dividing the EDI by the reference dose of THg (0.3 µg/kg/day) (USEPA, 1989). If HQ>1, a noncarcinogenic effect is possible; however, if HQ≤1, no noncarcinogenic effect is observed.

RESULTS AND DISCUSSION

MERCURY LEVELS IN AGRICULTURAL SOIL

Table 1 summarizes the THg concentrations, pH, and SOM of the soil samples. The THg concentrations in the agricultural soil ranged from 21 to 170 µg/kg dw (n=42), and the mean and median were 76±36 µg/kg dw and 70 µg/kg dw, respectively. In particular, the THg concentrations ranged from 21 to 170 µg/kg dw (mean±standard deviation (SD)=91±42 µg/kg dw, n=12) in Manso Abore, 37–170 µg/kg dw (76±39 µg/kg, n=15) in Nweneso No. 1, and 34–120 µg/kg dw (65±24 µg/kg dw, n=14) in Nweneso No. 2 (Table 1). The mean Hg concentrations in this study were comparable to those in the agricultural soil around the ASGM community in Teberebe, Ghana (72 µg/kg dw) (Bortey-Sam et al., 2015) (Table 2), and higher than those in the soil collected from the reference site (35 µg/kg dw) in this study and from across Ghana (mean 24 µg/kg, n=327) (Yevugah et al., 2023). However, the Hg concentration in this study was lower than those in soil collected from the ASGM area in Indonesia, the mine smelting area in Namibia, and the urban areas in Peru and China (Fang et al., 2011; Podolský et al., 2015; Quispe Aquin et al., 2022; Arrazy et al., 2023) (Table 2).

Table 1 Total mercury concentration (µg/kg dw) in agricultural soil and cacao crop samples

		Manso abore	Nweneso No.1	Nweneso No.2	Reference sitea
All farms soil	n	12b	15b	14b	1
	Mean	91	76	65	35c
	Median	83	69	65	–
	SD	42	39	24	–
	pH	6.0±0.5	6.4±0.3	6.8±0.6	8.1
	SOM (%)	6.8±2.7	6.9±2.5	7.2±2.3	12
Cacao farms soil	n	4	5	3	1
	Mean	115	116	87	35c
	Median	123	92	77	–
	SD	48	41	23	–
	pH	5.3±0.5	5.8±0.2	6.3±0.2	–
	SOM (%)	5.4±2.1	4.8±2.4	5.2±0.9	–
Cacao root	Mean	460	210	160	64
	Median	510	200	170	–
	SD	150	72	44	–
Leaves	Mean	260	130	160	57
	Median	170	130	120	–
	SD	180	19	120	–
Pods	Mean	47	35	35	34
	Median	72	35	34	–
	SD	22	7.0	4.0	–
Beans	Mean	100	25	23	11
	Median	75	23	26	–
	SD	61	9.0	6.0	–

^a Samples from three sites were composited. ^b The number of soil samples included the number of soil samples collected from the cacao farm. ^c The same sample.

Table 2 Comparison of Hg concentration (µg/kg dw) in agricultural soil from different countries

Location	Characteristics	Mean	Median	Reference
Manso Abore, Ghana	ASGM	91	83	This study
Nweneso No. 1, Ghana	ASGM	76	69	This study
Nweneso No. 2, Ghana	ASGM	65	65	This study
Teberebe, Ghana	ASGM	72		Bortey-Sam et al. (2015)
Lebaksitu, Indonesia	ASGM	1237		Novirsa et al. (2019)
MandailingNatal District, Indonesia	ASGM	19000		Arrazy et al. (2023)
Darmali, Sudan	ASGM		57	Elwaleed et al. (2024)
Peru	ASGM	72600		Quispe Aquin et al. (2022)
Namibia	Mine smelting	390	20	Podolský et al. (2015)
Zambia	Mine smelting	20	10	Podolský et al. (2015)
Renca, Central-northern Chile	Coal-fired power plant	355		Pérez et al. (2019)
Peru	Urban area	9500		Quispe Aquin et al. (2022)
Wuhu, China	Urban area	207		Fang et al. (2011)
San Luis Potosi, Mexico	Urban area	450		Perez-Vazquez et al. (2015)

The concentration of bioavailable Hg in the cacao farm soil ranged from 1.2 to 9.2 μg/kg dw, with a mean concentration of 6.0 μg/kg dw, which accounts for 1.1% to 15.5% of the THg concentration in the soil. The mean bioavailable Hg concentration was comparable to 3.8 μg/kg (0.28%–6.44% of THg) in agricultural soil in a previous study (Fang et al., 2011).

HORIZONTAL DISTRIBUTION OF THg IN AGRICULTURAL SOIL SAMPLES

The spatial distributions of THg concentrations in the three communities are shown in Fig. 2. Higher THg concentrations were observed around the ASGM site than at other sites in Nweneso No. 1, but this was not observed in Nweneso No. 2 or in Manso Abore. The difference in the THg content in the agricultural soil may be affected by the distance from the THg source. Previous studies have suggested a negative correlation between THg levels and the distances from point sources, such as the amalgam burning area, to the sampling points (Carpi, 1997; Novirsa et al., 2019; Coker et al., 2022; Arrazy et al., 2023; Thi Quynh et al., 2024). In this study, the burning site of the amalgam that included Hg in the communities could not be specified, and the relationship between the THg levels in the agricultural soil and the distance from the burning site was not clear.

Fig. 2 Spatial distribution of THg in agricultural soil in the three communities. Map created by the authors using data from © OpenStreetMap contributors, available under the Open Database License (https://www.opendatacommons.org/licenses/odbl/1.0/)

ASSESSMENT OF THg POLLUTION IN AGRICULTURAL SOIL

The CF ranged from 0.6 to 5.0 (Fig. 3(a)) and varied widely between the sampling sites, indicating a contamination level from low (CF<1) to considerable (3≤CF<6). The highest mean CF was found in Manso Abore, followed by Nweneso No. 1 and Nweneso No. 2. Seven sites in the study area exhibited considerable contamination. The mean CF was 2, and this value corresponded to moderate contamination.

Fig. 3 Comparison of the ASGM communities in terms of contamination factors (CF) and the potential ecological risk index (PERI) for agricultural soils

MA=Manso Abore, NW No. 1=Nweneso No. 1 and NW No. 2=Nweneso No. 2

The PERI ranged from 23 to 199 (Fig. 3(b)). The mean PERI value in the study area soil was 90, demonstrating that THg from agricultural soil poses considerable potential ecological risk. The mean value in the Manso Abore group was higher than that in the Nweneso No. 1 and Nweneso No. 2 groups. Two sites in Manso Abore and Nweneso No. 1 showed a high potential ecological risk (160≤Er<320). The mean PERI in this study was higher than that of Tarkwa, which was 61 (Bortey-Sam et al., 2015), but lower than that of a similar study conducted in Obuasi, Ghana, which was 469 (Addai-Arhin et al., 2023).

CONCENTRATION OF THg IN CACAO ROOTS, LEAVES, PODS, AND BEANS

Table 1 shows the mean and SD of the THg concentrations in cacao roots, leaves, pods, and beans. The mean THg concentration in cacao beans was 49±47 μg/kg (n=12). In previous studies, the THg concentration in cacao beans varied considerably from the detection limit (Vītola and Ciproviča, 2016; Assa et al., 2018) to 13,990 μg/kg (Anyimah-Ackah et al., 2021). There are no regulations or guidelines for THg in cacao beans, but such a large variation in the concentration suggests that the Hg in cacao beans should be monitored. The large variation in the THg concentration may represent the difference in the Hg load originating from anthropogenic activities such as ASGM and coal burning in the cultivation areas, and differences in accumulation by plant species and soil characteristics such as pH, SOM, and cation exchange capacity (Yu et al., 2018).

The THg concentrations in the cacao crop samples showed the following trend: roots>leaves>pods>beans (Fig. 4). Higher concentrations of Hg in roots and leaves than in other plant parts have been reported (Zhao and Duo, 2015; Jameer Ahammad et al., 2018). Plants are known to take up atmospheric gaseous mercury (Hg⁰) through their foliage by stomatal and cuticular uptake and then transport Hg through leaf tissues and translocate Hg to woody tissues by phloem transport (Zhou et al., 2021). In addition, although plants take up Hg from the soil, the transport of Hg through the root tissues to the xylem is minimal (Zhou et al., 2021). Roots play the role of a barrier in supplying metals to other aerial parts of plants (Rosales-Huamani et al., 2020). Ionic forms, such as Hg²⁺, which are water soluble, are retained by the cell wall components, limiting their entry into the plant (Tiodar et al., 2021). Hg²⁺ can bind to organic acids, such as citrate and malate, or to sulfur-rich structural proteins in the root apoplast cell wall (Carrasco-Gil et al., 2013; Montiel-Rozas et al., 2016; Tiodar et al., 2021). According to the Eh–pH diagram for an Hg–O–H–S–Cl system, the thermodynamically stable species of Hg in soil at around pH 5 vary from Hg⁰ to Hg₂Cl₂, Hg₂²⁺, HgCl₄²⁻, and Hg²⁺ with an increasing Eh value (oxidation–reduction potential) (Arbestain et al., 2009). In this study, we were not able to identify the Hg species in the soil, but the retention of Hg in the root suggests that these inorganic or ionic forms accumulate in the root.

Fig. 4 Comparison of Hg concentrations in agricultural soil and cacao crop samples. Different letters indicate significant differences between the media (p<0.05)

RELATIONSHIP BETWEEN THg CONCENTRATIONS IN SOIL AND CACAO SAMPLES

Table 3 presents the correlation matrix for the THg concentrations in the soil and cacao crop samples. A significant positive correlation was observed between the cacao crop samples (p<0.05), but not between the soil samples and cacao crop samples (p>0.05). The correlation coefficients between the bioavailable Hg in the soil and the cacao crop parts were slightly stronger than the THg concentration in the soil and the cacao crop parts. However, only cacao leaves showed a statistically significant correlation with the available Hg concentration in the soil (Table 3).

Table 3 Spearman’s rank correlation coefficient between Hg concentrations in soil and cacao crop samples

	Soil	Soilbioavailable	Beans	Leaves	Root	Pods
Soil	1
Soilbioavailable	0.13	1
Beans	−0.31	0.35	1
Leaves	0.34	0.57*	0.35	1
Root	0.12	0.43	0.68*	0.60*	1
Pods	0.03	0.38	0.80*	0.50*	0.87*	1

*  Correlation is significant at p<0.05.

The PCA biplot is shown in Fig. 5. In PC1, the THg concentrations in the cacao pod and bean showed negatively high factor loadings, and the pH in the soil showed a positive factor loading. This result suggests that the high THg concentration in cacao beans and pods is related to the low pH of the soil. The samples collected in Manso Abore (MA9, MA10, and MA12) showed lower pH (4.6–5.3) and higher THg concentrations in the cacao pods and beans than the samples collected at Nweneso No. 1 and No. 2. These samples were located in quadrants II and III of Fig. 5. However, all samples from Nweneso No. 1 and No. 2, which have moderate pH (5.4–6.5), are located in quadrants I and IV (Fig. 5). Only the SOM showed a negative high PC loading in PC2, suggesting that PC2 is related to the amount of organic matter in the soil. SOM is the main factor in Hg adsorption in tropical soil (Soares et al., 2015), but it was not associated with the transfer of Hg to the cacao pods and beans in this study. Interestingly, the factor loading of the bioavailable Hg concentrations in the soil corresponded well with the roots and the leaves. This result suggests that bioavailable Hg concentrations in the soil may be transferred to the roots and leaves, but not to the cacao pods.

Fig. 5 PCA biplot of the agricultural soil and cacao samples. Dots show the PC scores of samples, and vectors show the loadings of variables. The abbreviations MA, NW1, and NW2 indicate samples collected from Manso Abore, Nweneso No. 1, and Nweneso No. 2, respectively

BIOACCUMULATION AND TRANSFER FACTORS IN CACAO CROP SAMPLES

The BAF from the agricultural soil to the roots for all samples ranged from 1.1 to 10.7 (with a mean of 3.1), indicating a high accumulation in the roots from the contaminated soil (BAF>1) (Table 4). Using the Friedman ANOVA test, significant differences were observed between the BAF from the soil to the root and from the root to the aerial parts (leaves, pods, and beans) (p<0.05). However, no significant difference was observed in the TF from the root to the aerial parts (p>0.05). With respect to the TF from the root to the aerial parts, the mean TF was 0.73 in the leaves, and only two of the leaves had a TF>1, with the highest TF value of 1.7. The TF of 0.17 in the cacao beans was comparable to that of the pods, which was 0.19 (p>0.05). Generally, the cacao beans and pods had a TF below 1, indicating a possible low THg transfer from the root to the beans and the pod. Other studies on THg uptake and translocation by indigenous plants have also reported higher transfer in the root than in the aerial parts (Jameer Ahammad et al., 2018). Different results were reported in the cocoyam plant (Colocasia esculenta), which demonstrated higher transfer from the root to the leaves (0.39) than from the soil to the root (0.20) (Asare et al., 2021). Lower transfers of THg from agricultural soil to cassava tuber (0.039) and plantain (0.011) were obtained around the Tarkwa area, Ghana, than in the current study (Bortey-Sam et al., 2015).

Table 4 Bioaccumulation and transfer factors in cacao crop samples

	Bioaccumulation factor	Transfer factor
	Soil-root	Root-leaves	Root-beans	Root-pods
Reference site	1.83	0.89	0.17	0.55
Manso abore	5.4±4.2a	0.53±0.27b	0.17±0.07b	0.23±0.14b
Nweneso No. 1	1.9±0.78a	0.93±0.60b	0.23±0.12b	0.12±0.09b
Nweneso No. 2	2.0±0.93a	0.66±0.30b	0.18±0.06b	0.13±0.06b
All areas	3.1±2.7a	0.73±0.40b	0.17±0.10b	0.19±0.1b

^a,b  Different letters indicate significant differences between the media (p<0.05, ANOVA Tukey test).

ASSESSMENT OF HUMAN HEALTH RISK OF THg FROM CACAO PRODUCTS

The EDI of THg in processed products, i.e., chocolate from cacao beans, considering the worst-case scenario (100% maximum THg levels in chocolate transferred to the finished product), is presented in Table 5. The EDI was assessed using cocoa consumption per capita in Ghana, which is 0.52 kg/year. The estimated EDI was lower than the tolerable intake limit of THg in foodstuffs, which is 1.0 µg/kg/day (Salama, 2019). The HQ was estimated to assess the noncarcinogenic health risk posed by ingesting the THg in cacao beans. The HQ values for every sample were less than 1, indicating that Ghanaians consuming cacao beans in the study areas had no noncarcinogenic effect; therefore, cacao bean consumption was safe.

Table 5 Health risk assessment of Hg in cacao beans

Cacao beans origin	Max conc. (μg/kg)	Estimated daily intake (μg/kg/day)	HQ
Manso abore	188	0.004	0.01
Nweneso No. 1	40	0.0008	0.003
Nweneso No. 2	27	0.0005	0.002

CONCLUSIONS

This study focused on cacao plants in three ASGM communities in Ghana that used Hg in an amalgamation process and uncovered the relationship between the THg concentrations of cacao and soil and the environmental and health risks. The mean THg concentration in the soil samples was 76±36 μg/kg dw, which was higher than that in the soil samples from other parts of Ghana. The pollution indices (CF and PERI) indicated environmental contamination and related ecological risk in the study area. Although the THg concentrations in the agricultural soil accumulated approximately three times in the cacao root, the transfer of Hg from the root to the aerial parts of the plant, such as leaves, pods, and beans, was minimal. Therefore, the human health risk assessment indicated that there is no possibility of noncarcinogenic health effects of THg following the consumption of cacao beans from the study area. Thus, this study found little accumulation of THg from the soil to the cacao beans. However, the high concentration of THg in the agricultural soils of the study area may accumulate in other plants and may pose an ecological risk to wildlife, as suggested by the PERI. Therefore, strict measures must be taken to reduce the release of Hg and to remediate the affected area.

ACKNOWLEDGMENTS

The authors acknowledge the support of the Kumamoto Prefectural Government and the Prefectural University of Kumamoto in Japan for this study.

DATA AND MATERIAL AVAILABILITY

The data and materials used in the current research are available upon request from the corresponding author (Email: [e-mail]).

COMPETING INTERESTS

The authors have no competing interests to declare.

FUNDING

This study was funded by the Kumamoto Prefectural Government and the Prefectural University of Kumamoto, Japan, through the International postgraduate scholarship for research on mercury.

AUTHOR CONTRIBUTIONS

P.A.P.: Conceptualization, research design, sampling, data collection, sample analysis, statistical analysis, and writing of the first draft of the manuscript. B.M.: Sampling, writing-review, and editing. S.A.A.: Research design, sampling, writing-review, and editing. A.E.: review, sample analysis, and editing. T. A.: Review and editing. H.J.: Writing-review and editing. T.A.: Research design, writing-review and editing. Y.I.: Conceptualization, writing-review, and editing. J.K.: Research design, writing-review, editing, and supervision.

REFERENCES

•  Addai-Arhin,  S.,  Novirsa,  R.,  Jeong,  H.,  Phan,  Q.D.,  Hirota,  N.,  Ishibashi,  Y.,  Shiratsuchi,  H.,  Arizono,  K., 2023. Mercury waste from artisanal and small-scale gold mining facilities: a risk to farm ecosystems—A case study of Obuasi, Ghana. Environ. Sci. Pollut. Res. 30(2), 4293–4308. doi: 10.1007/s11356-022-22456-4.
•  Ahoa,  E.,  Kassahun,  A.,  Tekinerdogan,  B.,  Verdouw,  C., 2021. Analyzing and designing business processes in the Ghana cocoa supply chain to support inclusiveness. Sustainability (Switzerland) 13(22), 12440. doi: 10.3390/su132212440.
•  Amankwaah,  D.,  Asante Nnuro,  W.A.,  Awudza,  J.,  Afful,  S., 2015. Determination of heavy metals in cocoa beans from some major cocoa-growing regions in Ghana. Ann. Food Sci. Technol. 16(1), 225–235. https://afst.valahia.ro/wp-content/uploads/2022/09/s03_w01_full_2015.pdf
•  Anyimah-Ackah,  E.,  Ofosu,  I.W.,  Lutterodt,  H.E.,  Darko,  G., 2021. Health Benefit: Risk Assessment of Trace and Essential Elements Found in Cocoa Beans and Derived Products. Chem. Africa 4(2), 299–312. doi: 10.1007/s42250-020-00214-8.
•  Arbestain,  M.C.,  Rodríguez-Lado,  L.,  Bao,  M.,  Macías,  F., 2009. Assessment of Mercury-Polluted Soils Adjacent to an Old Mercury-Fulminate Production Plant. Appl. Environ. Soil Sci. 2009, 387419. doi: 10.1155/2009/387419.
•  Arham,  Z.,  Asmin,  L.O., Rosmini, Nurdin, M., 2017. Heavy metal content of cocoa plantation soil in East Kolaka, Indonesia. Orient. J. Chem. 33(3), 1164–1170. doi: 10.13005/ojc/330314.
•  Arrazy,  S.,  Addai-Arhin,  S.,  Jeong,  H.,  Novirsa,  R.,  Wispriyono,  B.,  Agusa,  T.,  Ishibashi,  Y.,  Kobayashi,  J., 2023. Spatial distribution and human health risks of mercury in the gold mining area of Mandailing Natal District, Indonesia. Environ. Monitor. Contam. Res. 3, 33–42. doi: 10.5985/emcr.20230003.
•  Asare,  D.A.,  Anderson,  E.,  Amosah,  M.E.,  Ntiri-Bekoh,  R.,  Ofori-Sarpong,  G., 2021. Assessing the Toxicity of Colocassia Esculenta (Cocoyam Plant) Grown on Mercury Contaminated Soil. Ghana Mining J. 21(2), 46–54. doi: 10.4314/gm.v21i2.6.
•  Assa,  A.,  Noor,  A.,  Yunus,  M.R., Misnawi, Djide, M.N., 2018. Heavy metal concentrations in cocoa beans (Theobroma cacao L.) originating from East Luwu, South Sulawesi, Indonesia. J. Phys.: Conf. Ser. 979, 012011. doi: 10.1088/1742-6596/979/1/012011.
•  Barraza,  F.,  Schreck,  E.,  Uzu,  G.,  Lévêque,  T.,  Zouiten,  C.,  Boidot,  M.,  Maurice,  L., 2021. Beyond cadmium accumulation: Distribution of other trace elements in soils and cacao beans in Ecuador. Environ. Res. 192, 110241. doi: 10.1016/j.envres.2020.110241.
•  Bortey-Sam,  N.,  Nakayama,  S.M.M.,  Akoto,  O.,  Ikenaka,  Y.,  Baidoo,  E.,  Mizukawa,  H.,  Ishizuka,  M., 2015. Ecological risk of heavy metals and a metalloid in agricultural soils in Tarkwa, Ghana. Int. J. Environ. Res. Public Health 12(9), 11448–11465. doi: 10.3390/ijerph120911448.
•  Carpi,  A., 1997. Mercury from Combustion Sources: A Review of the Chemical Species Emitted and Their Transport in the Atmosphere. Water, Air, Soil Pollut. 98, 241–254. doi: 10.1023/A:1026429911010.
•  Carrasco-Gil,  S.,  Siebner,  H.,  Leduc,  D.L.,  Webb,  S.M.,  Millan,  R.,  Andrews,  J.C.,  Hernandez,  L.E., 2013. Mercury localization and speciation in plants grown hydroponically or in a natural environment. Environ. Sci. Technol. 47, 3082–3090.
•  Coker,  E.A.,  Nkuah,  B. C.,  Amoanimaah,  S. A.,  Oppong,  J. B.,  Gyamfi,  O.,  Ansah,  E.,  Ababio,  K.A.,  Wemegah,  D.D.,  Borquaye,  L.S.,  Darko,  G., 2022. Human exposure to mercury in the atmosphere and soils in Konongo: an age-old mining centre in the Ashanti Region of Ghana. Environ. Geochem. Health. 45(6), 3555–3565. doi: 10.1007/s10653-022-01441-3.
•  Elwaleed,  A.,  Jeong,  H.,  Abdelbagi,  A.H.,  Quynh,  N.T.,  Agusa,  T.,  Ishibashi,  Y.,  Arizono,  K., 2024. Human Health Risk Assessment from Mercury-Contaminated Soil and Water in Abu Hamad Mining Market, Sudan. Toxics 12(2), 112. doi: 10.3390/toxics12020112.
• Environmental Protection Agency, Ghana, 2018. Minamata convention on mercury initial assessment report for Ghana. https://www.undp.org/sites/g/files/zskgke326/files/migration/gh/UNDP_GH_SDC_MIA_REPORT_2018-1_0.pdf (accessed 17 May 2025)
•  Fang,  F.,  Wang,  H.,  Lin,  Y., 2011. Spatial distribution, bioavailability, and health risk assessment of soil Hg in the Wuhu urban area, China. Environ. Monitor. Assess. 179(1–4), 255–265. doi: 10.1007/s10661-010-1733-8.
• GCB Strategy & Research Dept., 2023. Cocoa Sector Analysis 2023. https://www.gcbbank.com.gh/research-reports/sector-industry-reports/356-cocoa-industry-in-ghana-2023/file (accessed 17 May 2025)
•  Gyamfi,  O.,  Wireko-Gyebi,  R.S.,  Ansah,  E.,  Sorenson,  P.B.,  King,  R.S.,  Nkansah,  M.A.,  Bak,  J.L.,  Darko,  G., 2022. Assessment and Awareness of Health Risks Posed by Mercury in Artisanal Gold Mining in the Ashanti Region of Ghana. Chem. Africa 5(5), 1765–1775. doi: 10.1007/s42250-022-00453-x.
•  Hakanson,  L., 1980. An ecological risk index for aquatic pollution control—A sedimentological approach. Water Res. 14(8), 975–1001. doi: 10.1016/0043-1354(80)90143-8.
•  Jameer Ahammad,  S.,  Sumithra,  S.,  Senthilkumar,  P., 2018. Mercury uptake and translocation by indigenous plants. Rasayan J. Chem. 11(1), 1–12. doi: 10.7324/RJC.2018.1111726.
•  Kithure,  J.G.N.,  Hudumu,  J.C., 2023. Determination of Heavy Metals in Cocoa Products Consumed in Nairobi County Kenya. Int. J. Latest Technol. Eng., Manag. Appl. Sci. 12(7), 96–105. doi: 10.51583/IJLTEMAS.2023.12709.
•  Liu,  J.,  Wang,  W.,  Peng,  A., 1997. Study on the mercury species of Beijing surface soils and sediments. Environ. Chem. 16(2), 171–177 (in Chinese).
•  Mantey,  J.,  Nyarko,  K.B.,  Owusu-Nimo,  F.,  Awua,  K.A.,  Bempah,  C.K.,  Amankwah,  R.K.,  Akatu,  W.E.,  Appiah-Effah,  E., 2020. Influence of illegal artisanal small-scale gold mining operations (galamsey) on oil and grease (O/G) concentrations in three hotspot assemblies of the Western Region, Ghana. Environ. Pollut. 263, 114251. doi: 10.1016/j.envpol.2020.114251.
•  Montiel-Rozas,  M.M.,  Madejón,  E.,  Madejón,  P., 2016. Effect of heavy metals and organic matter on root exudates (low molecular weight organic acids) of herbaceous species: An assessment in sand and soil conditions under different levels of contamination. Environ. Pollut. (Barking, Essex: 1987) 216, 273–281. doi: 10.1016/j.envpol.2016.05.080.
•  Mounicou,  S.,  Szpunar,  J.,  Andrey,  D.,  Blake,  C.,  Lobinski,  R., 2003. Concentrations and bioavailability of cadmium and lead in cocoa powder and related products. Food Addit. Contam. 20(4), 343–352. doi: 10.1080/0265203031000077888.
•  Novirsa,  R.,  Dinh,  Q.P.,  Jeong,  H.,  Fukushima,  S.,  Ishibashi,  Y.,  Wispriyono,  B.,  Arizono,  K., 2019. Evaluation of mercury contamination in upland rice paddy fields around an artisanal small-scale gold mining area, Lebaksitu, Indonesia. J. Environ. Saf. 10(2), 119–125.
•  Olawale,  O.F.,  Abah,  M.A.,  Emmanuel,  O.P.,  Otitoju,  G.T.,  Abershi,  A.L.,  Temitope,  D.F.,  Andrew,  A.E.,  Abdulkadir,  S.,  John,  A., 2023. Risk assessment of the heavy metal content in yam tubers locally produced in selected local government areas of Taraba State, Nigeria. Asian J. Nat. Product Biochem. 21(1), 6–12. doi: 10.13057/biofar/f210102.
•  Osman,  N.,  Tennyson Afele,  J.,  Nimo,  E.,  Gorleku,  D.O.,  Ofori,  L.A.,  Abunyewa,  A.A., 2022. Assessing the Impact of Illegal Small-Scale Mining (Galamsey) on Cocoa Farming and Farmer Livelihood: A Case Study in the Amansie West District of Ghana. Pelita Perkebunan (a Coffee and Cocoa Research Journal) 38(1), 70–82. doi: 10.22302/iccri.jur.pelitaperkebunan.v38i1.496.
•  Pérez,  P.A.,  Hintelmann,  H.,  Lobos,  G.,  Bravo,  M.A., 2019. Mercury and methylmercury levels in soils associated with coal-fired power plants in central-northern Chile. Chemosphere 237, 124535. doi: 10.1016/j.chemosphere.2019.124535.
•  Perez-Vazquez,  F.J.,  Flores-Ramirez,  R.,  Ochoa-Martinez,  A.C.,  Orta-Garcia,  S.T.,  Hernandez-Castro,  B.,  Carrizalez-Yañez,  L.,  Pérez-Maldonado,  I.N., 2015. Concentrations of persistent organic pollutants (POPs) and heavy metals in soil from San Luis Potosí, México. Environ Monit. Assess 187(1), 4119. doi: 10.1007/s10661-014-4119-5.
•  Podolský,  F.,  Ettler,  V.,  Šebek,  O.,  Ježek,  J.,  Mihaljevič,  M.,  Kříbek,  B.,  Sracek,  O.,  Vaněk,  A.,  Penížek,  V.,  Majer,  V.,  Mapani,  B.,  Kamona,  F.,  Nyambe,  I., 2015. Mercury in soil profiles from metal mining and smelting areas in Namibia and Zambia: distribution and potential sources. J. Soils Sediments 15(3), 648–658. doi: 10.1007/s11368-014-1035-9.
•  Quispe Aquin,  R.,  Malone,  A.,  Smith,  N.M.,  García Zúñiga,  F.F., 2022. Perceptions and realities of mercury contamination in a Peruvian artisanal and small-scale gold mining (ASGM) community. Environ. Res. 214, 114092. doi: 10.1016/j.envres.2022.114092.
•  Rajaee,  M.,  Sánchez,  B.N.,  Renne,  E.P.,  Basu,  N., 2015. An investigation of organic and inorganic mercury exposure and blood pressure in a small-scale gold mining community in Ghana. Int. J. Environ. Res. Public Health 12(8), 10020–10038. doi: 10.3390/ijerph120810020.
• R Core Team 2025. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/
•  Rosales-Huamani,  J.A.,  Breña-Ore,  J.L.,  Sespedes-Varkarsel,  S.,  De la Cuba,  L.H.,  Centeno-Rojas,  L.,  Otiniano-Zavala,  A.,  Andrade-Choque,  J.,  Valverde-Espinoza,  S.,  Castillo-Sequera,  J.L., 2020. Study to Determine Levels of Cadmium in Cocoa Crops Applied to Inland Areas of Peru: “The Case of the Campo Verde-Honoria Tournavista Corridor.” Agronomy 10, 1576. doi: 10.3390/agronomy10101576.
•  Salama,  A.K., 2019. Health risk assessment of the heavy metal content in cocoa and chocolate products sold in Saudi Arabia. Toxin Rev. 38(4), 318–327. doi: 10.1080/15569543.2018.1471090.
•  Soares,  L. C.,  Egreja Filho,  F. B.,  Linhares,  L. A.,  Windmoller,  C. C.,  Yoshida,  M. I., 2015. Accumulation and oxidation of elemental mercury in tropical soils. Chemosphere 134, 181–191. doi: 10.1016/j.chemosphere.2015.04.020.
•  Tetteh,  S.,  Golow,  A.A.,  Essumang,  D.K.,  Zugle,  R., 2010. Levels of mercury, cadmium, and zinc in the topsoil of some selected towns in the Wassa West District of the Western region of Ghana. Soil Sediment Contam. 19(6), 635–643. doi: 10.1080/15320383.2010.515626.
•  Thi Quynh,  N.,  Jeong,  H.,  Elwaleed,  A.,  Nugraha,  W.C.,  Arizono,  K.,  Agusa,  T.,  Ishibashi,  Y., 2024. Spatial and Seasonal Patterns of Mercury Accumulation in Paddy Soil around the Nam Son Landfill, Hanoi, Vietnam. Soil Syst. 8(1), 30. doi: 10.3390/soilsystems8010030.
•  Tiodar,  E.D.,  Văcar,  C.L.,  Podar,  D., 2021. Phytoremediation and Microorganisms-Assisted Phytoremediation of Mercury-Contaminated Soils: Challenges and Perspectives. Int. J. Environ. Res. Public Health 18(5), 2435. doi: 10.3390/ijerph18052435.
• United Nations Environment Programme, 2019. Global Mercury Assessment 2018. https://www.unep.org/resources/publication/global-mercury-assessment-2018 (accessed 17 May 2025)
• U.S. Environmental Protection Agency, 1989. Risk Assessment Guidance for Superfund Volume I Human Health Evaluation Manual (Part A). https://www.epa.gov/sites/default/files/2015-09/documents/rags_a.pdf (accessed 17 May 2025)
•  Vītola,  V.,  Ciproviča,  I., 2016. The Effect of Cocoa Beans Heavy and Trace Elements on the Safety and Stability of Confectionery Products. Rural Sustain. Res. 35(330), 19–23. doi: 10.1515/plua-2016-0003.
• World Health Organization, 2020. 10 chemicals of public health concern. World Health Organization. https://www.who.int/news-room/photo-story/detail/10-chemicals-of-public-health-concern (accessed 17 May 2025)
•  Yevugah,  L.L.,  Darko,  G.,  Bak,  J., 2021. Does mercury emission from small-scale gold mining cause widespread soil pollution in Ghana? Environ. Pollut. 284, 116945. doi: 10.1016/j.envpol.2021.116945.
•  Yevugah,  L.L.,  Osei Jnr,  E.M.,  Bak,  J.L.,  Nkansah,  M.A.,  Darko,  G., 2023. Geospatial Distribution of Mercury in Surface Soils Across Ghana. Chem. Africa 6, 3119–3129. doi: 10.1007/s42250-023-00681-9.
•  Yoon,  J.,  Cao,  X.,  Zhou,  Q.,  Ma,  L.Q., 2006. Accumulation of Pb, Cu, and Zn in native plants growing on a contaminated Florida site. Sci. Total Environ. 368(2–3), 456–464. doi: 10.1016/j.scitotenv.2006.01.016.
•  Yu,  H.,  Li,  J.,  Luan,  Y., 2018. Meta-analysis of soil mercury accumulation by vegetables. Scientific reports 8(1), 1261. doi: 10.1038/s41598-018-19519-3.
•  Zhao,  S.,  Duo,  L., 2015. Bioaccumulation of cadium, copper, zinc, and nickel by weed species from municipal solid waste compost. Polish J. Environ. Studies 24(1), 413–417. doi: 10.15244/pjoes/28960.
•  Zhou,  J.,  Obrist,  D.,  Dastoor,  A.,  Jiskra,  M. and  Ryjkov,  A., 2021. Vegetation uptake of mercury and impacts on global cycling. Nat. Rev. Earth Environ. 2, 269–284. doi: 10.1038/s43017-021-00146-y.