The Cavally River is the primary waterway that drains the Zouan-Hounien department ( Figure 1 ). Rising in Guinea north of Mount Nimba at an altitude of 600 metres, it is 700 km long. The 700 km-long Cavally River serves as a natural border between Côte d’Ivoire and Liberia in its middle and lower reaches [14] . The Zouan-Hounien department has a mountain climate, characterized by two seasons: a rainy season from March to October, and a dry season from November to February. The average annual rainfall is 1.87 mm (1970-2015). January is the least rainy month, with an average rainfall of less than 15 mm. Conversely, September is the wettest month, with an average rainfall of 236 mm and a maximum discharge of 196.01 m³∙s⁻¹. The fish species selected for this study was the catfish, scientific name Clarias nigrodigitatus ( Figure 2 ), caught in the Cavally River during the low-water period.

Fish samples were obtained during a single sampling campaign from local fishermen. Muscles were utilized as the fish components employed for TME determination, as they represent the edible portions of the fish and function as a circumstantial biological indicator, given their relative mass in comparison to other organs [15] . A total of 27 fish samples were collected, stored in a cooler containing ice cubes, and sent to the Central Laboratory for Food Hygiene and Agro-industry (LCHAI) of the National Laboratory for Support to Agricultural Development (LANADA) for chemical analysis in Abidjan. Upon arrival at the designated laboratory, the fish samples were stored in the freezer until the specific parts required for the TME assay were extracted.

In the laboratory, the fish were measured from mouth to caudal fin tip (total length) and to the fork (standard length) using an ichthyometer, and then weighed on a 0.001 g precision electronic balance (total weight). For samples whose mass exceeded 5 kg, a pesola or load cell was employed for mass measurements. Subsequent to the collection of the aforementioned measurements, a 5 g sample of skinless dorsal muscle was obtained from each specimen. The fresh part was ground in an agate mortar and then placed in Teflon tubes, to which 6 ml of nitric acid (HNO3 68%) was added and closed with watch glasses. Subsequent to solubilisation, the tubes were placed in a sand bath at a gradual rate of heating to 130˚C, until the complete evaporation of the acid. The resulting pellets were then dissolved in hot distilled water. Following the cooling process, each acid-etch solution was transferred volumetrically to a 25 ml flask. Following homogenisation, the solution was filtered through filter paper. Concurrently, a 6 ml solution of 68% HNO₃ was prepared under identical conditions. With the exception of mercury, the presence of TME was detected by graphite furnace atomic absorption spectrometry (Perkin-Elmer spectrophotometer, model 3030).

The intensity of contamination in fish is estimated by comparing the mean value of TME concentrations in each muscle sample with the guide values for the maximum limit concentrations of trace elements in fish [16] ( Table 1 ).

The assessment of metal contamination in fish was carried out by calculating the bioconcentration factor (BCF), a parameter linking the concentration of the contaminant in the organism of the species to that of the environment (water, sediment, soil, and air). It is used to describe the transfer of a contaminant from the biotope to the organism. It is expressed as the ratio of the concentration of the contaminant (inorganic parameter) in the organism (fish) to that of the contaminant in the biotope (water and sediment) [17] . The higher the ratio, the more intense the bioconcentration of TMEs in the organism [18] . Its formula is given by Equation (1).

Both concentrations are expressed in mg∙kg⁻¹. When BCF > 1, there is bioconcentration of the TME in the fish and therefore its transfer from the biotope to the fish.

$\text{FBC}=\frac{\text{TME concentration in organism}}{\text{TME concentration in water or sediment}}$ (1)

The health risk assessment method was developed by the U.S. National Research Council [19] . Adverse effects on human health resulting from environmental pollution are defined as the probability of such effects arising [20] [21] . It is a method for calculating the health risks to which a population exposed to a particular pollution of anthropogenic or natural origin is subject [22] . In the present study, the research question was posed and subsequently resolved in two distinct age demographics: namely, children (under 15 years of age) and adults (over 15 years of age). The application process is comprised of four distinct stages:

$\text{EDI}=\frac{\text{C}\times \text{IR}\times \text{ED}\times \text{EF}}{\text{BW}\times \text{AT}}\times {10}^{-3}$(2)

where EDI = Estimated Daily Intake (mg∙kg⁻¹∙day⁻¹).

This involves calculating the hazard quotient (HQ) for threshold effects. It is obtained from Equation (3) according to [23] .

$\text{HQ}=\frac{\text{EDI}}{\text{RfD}}$(3)

When HQ < 1, the probability of a toxic effect occurring is negligible. Conversely, when HQ > 1, the possibility of a toxic effect cannot be discounted.

In the context of this study, a hypothetical scenario has been postulated in which an individual (child or adult) is assumed to consume fish on a daily basis for a period of seven days. It is further postulated that the average annual quantity of fish ingested by a child will be equivalent to that of an adult [24] . Furthermore, no amount of the metal measured in the muscle will be considered lost or amplified during the cooking process, and therefore the ingestion dose will be equal to the contaminant dose absorbed [12] .

The carcinogenic risk is defined as the probability that a person will develop cancer due to exposure to a particular contaminant or a set of contaminants in the environment [25] . It is given by the ratio of the daily exposure dose to the carcinogenic slope factor (CSF) according to Equation (4) [25] [26] .

$\text{Carcinogenic risk}=\text{EDI}\times \text{CSF}$(4)

The cancer slope factors for Pb, Cd, and As are 0.0085, 0.63, and 1.5 mg/kg/day, respectively [2 5] [27] .

Risks were categorized as negligible (CR < 10⁻⁶), an acceptable range (10⁻⁶ ≤ RI < 10⁻⁴) and unacceptable (RI > 10⁻⁴) based on the calculated values [25] [27] [ 28].

The mean concentrations of TME in fish muscle are presented in Table 3 . The mercury concentrations ranged from 0 to 0.06 mg∙kg⁻¹, with an average of 0.023 ± 0.01 mg∙kg⁻¹. It is evident that all 27 fish samples have concentrations below the limit value of 0.5 mg∙kg⁻¹ set by the Food and Agriculture Organization (FAO) and the World Health Organization (WHO) [16] . This limit value is considered permissible for fish. Lead concentrations ranged from 0 to 0.46 mg∙kg⁻¹, with a mean value of 0.25 ± 0.12 mg∙kg⁻¹. In 29.62% of samples (8/27), lead concentrations exceeded the permissible limit of 0.3 mg∙kg⁻¹. Cadmium concentrations ranged from 0 to 0.29 mg∙kg⁻¹, with an average of 0.011 ± 0.004 mg∙kg⁻¹. A total of 6 fish out of 27 (22.22%) had concentrations above the permissible limit of 0.025 mg∙kg⁻¹ for fish according to [16] . For arsenic, concentrations ranged from 0 to 0.40 mg∙kg⁻¹, with an average of 0.09 ± 0.01 mg∙kg⁻¹. All fish sampled had arsenic concentrations in line with recommended standards. Iron concentrations were highest, ranging from 5.56 to 41.36 mg∙kg⁻¹, with an average of 17.07 ± 5.36 mg∙kg⁻¹. However, all the fish analyzed (27/27), i.e. 100%, had Fe concentrations below the required standard of 43 mg∙kg⁻¹ for fish. These fish are therefore fit for human consumption. TME are abundant in fish sampled in Cavally River waters in the following order: Fe ≻ Pb ≻ As ≻ Hg ≻ Cd.

The results of the calculated bioconcentration factors for TMEs in water and sediments are reported in Table 4 . This table shows that the bioconcentration factors relative to water for mercury, lead, cadmium, arsenic, and iron are greater than one (BCF > 1). These values reflect an accumulation of trace metal elements in fish muscles from water. Conversely, the BCF relative to sediment for all trace metal elements is less than one (BCF < 1), showing that fish do not concentrate TMEs in their muscles from sediments.

The results of the estimated daily intake (EDI), hazard quotient (HQ), and carcinogenic risk (CR) calculated for children and adults are shown in Table 5 . EDI values for elements in children are lower than those in adults. Iron is the main element contributing to the risk of fish ingestion, with mean values well above those of the other elements. The estimated hazard quotient for all TME is less than 1 for both children and adults (HQ < 1). This indicates that there is likely no risk from eating catfish caught in the area. However, four out of eleven fish (4/11) recorded an HQ (As) greater than unity in children when considered individually. Arsenic is the element with the highest non-carcinogenic risk value in both children and adults. All cancer risk values obtained for these three metals are within the acceptable limit for cadmium and arsenic (10⁻⁶ ≤ CR < 10⁻⁴) and not significant for lead (CR < 10⁻⁶) in children and adults. These values show that long-term consumption of fish does not pose a cancer risk to the population.