The study was conducted in three communities that engaged in extensive farming along the banks of Lake Bosomtwe as shown in Figure 1. The Lake is located in the Bosomtwe-Atwima-Kwawoma District of the Ashanti Region of Ghana. The District lies within latitude 6˚43' North and longitude 1˚46' West.

The Lake Bosomtwe area is located within the equatorial zone with a rainfall regime typical of the moist semi-deciduous forest zone of Ghana [9]. The lake almost covers all the flat-leveled low lands between the high lands of the Bosomtwe basin and the remaining little flat-leveled low lands of the basin are used by local inhabitants to carry out subsistence agriculture. Most of the farmers are also involved in other cash crop farming such as cocoa, oil palm and citrus plantations which are done on a relatively small scale [9].

Field and laboratory techniques were used in this study. Purposive sampling technique was used to sample water from the selected communities for laboratory analysis. Fifty-four (54) lake water samples were collected from September,

2016 to February 2017 to the laboratory for analysis of pesticides and nutrient loads. Samples were analyzed using the procedures prescribed by American Public Health Association [10] with minor modifications. A structured questionnaire was used to assess the level of use of pesticides and fertilizers by inhabitants within the Bosomtwe basin. Using discriminative snowball which is also non-probability sampling technique, sixty (60) respondents, twenty (20) from each community were selected for interview to take inventory of pesticides and fertilizers application in the basin.

The water samples were filtered by passing each sample through a filter paper. A volume of 500 mL of each sample was extracted at room temperature using solid phase extraction method. A weak anion exchange solid extraction octadecyl C₁₈ cartridges were preconditioned with 5 mL of 0.1% NH₄OH in methanol followed by double distilled water. Water samples were passed through C₁₈ column cartridges through polypropylene tubing under vacuum at a high flow rate (10 mL per minute). A volume of 2 mL of methanol was used to elute the uncategorized organic compounds into a sample valve. Each extract was then dried using Na₂SO₄. Extracted samples were fortified with 20 μL of the 200 μg/L of mixed internal standards. The spiked extracts were subjected to nitrogen evaporation until the final volume required for analysis was obtained. The samples were analyzed using Gas Chromatograph Mass Spectrometer

Quantification and Limit of Detection of Pesticide

The residue levels of the pesticides were quantitatively determined by the external standard method using peak area. Measurement was carried out within the linear range of the detector following the procedures described by [11] [12]. The peak areas whose retention times coincided with the standards were extrapolated on their corresponding calibration curves to obtain the concentration. The limit of detection of the pesticides determined was based on the extract of the fortified samples that were serially diluted by a factor of two to give different concentrations. One out of each concentration that gave a response three times the standard deviation of the least fortified sample was noted

Linearity was evaluated in the 15 - 150 μg/L concentration range. Four standard concentrations (15, 30, 75 and 150 μg/L) were prepared and used to construct a calibration curve to check the linearity of the instrument response and also for quantification. The regression coefficients (R²) of the calibration curve were calculated by plotting area ratio against concentration. For most of the targeted compounds, regression coefficient was 0.99 indicating good linearity. Four blank samples were analyzed and the average concentrations of each analyte in the blanks were used to correct the concentration of the corresponding analyte in the test samples. Recovery test was conducted by spiking a mixture of native standards (150 μg/L) into deionized water and then extracted following the same procedures used for the real samples. The reproducibility of the method was then determined. The obtained recoveries and RSD values were largely satisfactory, indicating a fairly good precision for most of the analytes. The limit of detection (LOD) (i.e., concentration that yielded signal to noise ratio of ≥3) and limit of quantification (LOQ) (i.e., concentration that yielded signal to noise ratio of ≥10) for the various analytes were determined for each sample and the median LOD and LOQ values were reported as the threshold values for each analyte.

Nutrient Loads of the Lake Water

Phosphate and Nitrate were the main nutrients whose loads were determined following the standard procedures prescribed by reference [10].

The data obtained were analyzed using Statistical Product Service Solutions (SPSS version 20) for descriptive statistics and the results were presented as shown below.

Table1 shows the mean concentrations of pesticide loads of Lake Bosomtwe.

Source: field data, 2017.

Dichlorvos recorded a mean concentration of 0.11 ± 0.09 ppb with minimum and maximum concentrations of 0.03 ppb and 0.18 ppb respectively. The concentration of dichlorvos was a bit lower when compared with the WHO/USEPA guidelines of 5 ppb for surface water. “Reference [13] attributed movement of contaminants into the lake Bosomtwe to farming activities occurring in the basin.” According to [14], the toxicity of dichlorvos for freshwater and estuarine fish is moderate to high, and it does not bioaccumulate in fish. However, dichlorvos has the potential to induce altered immune response in fish [15].

Diazinon had a mean concentration of 0.28 ± 0.03 ppb during the study. “Reference [16] recorded lower concentration (0.03 ppb) of this same pesticide for surface waters in Dormaa West in Brong Ahafo Region of Ghana.” The mean concentration of diazinon was higher than the WHO 0.05 ppb and USEPA 0.04 ppb for surface water bodies. The detection of diazinon confirms the findings from the field survey, which revealed that diazinon, is an active ingredient of a pesticide with a trade name “Akate suro”, which is extensively used by cocoa farmers along the banks of the lake. Diazinon has been found to capably cause acute and chronic intoxication [17].

Chlorpyrifos and Methyl-Chlorpyrifos respectively had mean concentrations 0.16 ± 0.05 ppb and 0.19 ± 0.05 ppb. The mean concentrations of both chlorpyrifos and methyl-chlorpyrifos were higher when compared with the 0.08 ppb limit prescribed by the USEPA. Chlorpyrifos like other organophosphate insecticides acts on the nervous system of the parasites and other organisms such as fish, birds and mammals that come in contact with these pesticides. It inhibits acetylcholinesterase an enzyme that hydrolyses acetylcholine.

TPhP and Coumaphos respectively had mean concentrations of 0.16 ± 0.06 ppb and 0.10 ± 0.05 ppb. Coumaphos, TPhP and other organophosphate pesticides have been extensively reported by [6] in Fosu lagoon, Chemu lagoon, Korle lagoon all in Ghana. Coumaphos has been linked to cause so many disturbances in aquatic systems particularly lakes [18]. An increasing number of studies linked exposure to TPhP with reproductive and developmental toxicity, neurotoxicity, metabolic disruption, endocrine effects, and genotoxicity. TPhP has also been found to induce significant estrogenic activity. Organophosphates have been identified to be a contributing factor of eutrophication in water bodies particularly lakes [19].

Eight (8) problematic class of organochlorine pesticide residues were detected in the samples as shown in Table1. They include Lindane Dichlorobenzophenone, Heptachlorepoxide, Gamma chlordane, Alpha chlordane, Dieldrin, DDE and Endoslfan sulphate. It was observed from the analysis of the questionnaire that most of the organochlorine pesticides detected in the lake water were not actively used in the basin anymore but their concentration still persists. This could be in line with the assertion of [6] that OC’s are resistant to microbial and photolytic degradation, and are therefore persistent in the environment (soils and water) where they are applied. Similar concentrations of OC’s have been reported by [6] for Fosu, Korle and Chemu lagoons in Ghana.

Exposure to concentrations of organochlorine pesticides over a long period may eventually lead to a substantial body burden of toxic chemicals. Acute ingestion of aquatic products contaminated with organochlorine pesticides may cause a loss of sensation around the mouth; hypersensitivity to light, sound, and touch; dizziness, tremors, nausea, vomiting, nervousness, and confusion [20]. Population-based studies have revealed possible relations between the exposure to organochlorine pesticides and serious health effects including cardiovascular diseases, negative effects on the male reproductive system and on the nervous system, dementia, and also a possible increased risk for non-Hodgkin’s lymphoma [21].

Permetrin was the only pesticide identified under this group and a phenol type pesticide, σ-phenylphenol as shown in Table 1. The mean concentration recorded for Permetrin and σ-phenylphenol were 0.48 ± 0.30 ppb and 0.05 ± 0.05 ppb respectively. “Reference [16] reported similar concentration for surface water bodies in Dormaa West.” The mean concentration of Permetrin observed at the various sampled sites with detectable residues was found to be higher than the World Health Organization (WHO) Maximum Residue Limit of 0.05 ppb for Permetrin. It is reported that, exposure to synthetic pyrethroids can cause hyper-excitation, aggressiveness, incoordination, whole-body tremors, and seizures [16].

The nitrate levels recorded during the study ranged from 0.01 mg/L to 0.25 mg/L with a mean value of 0.15 ± 0.05 mg/L as shown in Table 2. The mean nitrate value recorded for the study was a bit lower when compared with the overall mean nitrate concentration of 1.3 mg/L recorded by [13] for Lake Bosomtwe. However, the concentration observed for this study (0.16 mg/L) by far fell in line with the nitrate levels obtained by [22] in Lake Volta (0.2 - 1.70 mg/L) and elsewhere in Nigeria (0.10 - 2.60 mg/L) by [23]. The level of nitrate in the lake was far below the W.H.O limit of 10 mg/L. Nitrate levels exceeding 5 mg/L in most cases are indicative of anthropogenic pollution which could be harmful to aquatic organisms and humans. The level of nitrate in the Bosomtwe could be attributed to the use of nitrogenous fertilizers along the bank of the lake by most farmers. “Reference [13] attributed nitrate levels in surface water to increase in surface runoff, bare areas and leaching of nitrogenous fertilizers from nearby farmlands.” A study by [24] concluded that, nitrogen fertilizer application contributed to an increase in nitrate levels in surface water through surface run-off. Fertilizers on farmlands could be sub-lethal to native fish species and other aquatic organisms [25].

The phosphate level ranged between 0.098 mg/L to 0.68 mg/L with a mean concentration of 0.40 ± 0.12 mg/L as shown in Table 2. The mean concentration of phosphate had marginally increased when compared with the mean values of 0.40 mg/L recorded by [13] in the same lake. The mean phosphate concentration for this study was slightly higher when compared with the values (0.031 mg/L) recorded by [26] in the same lake. In view of this, lake can be said to be polluted considering the fact that the minimum phosphate level for most uncontaminated lakes ranges from 0.01 - 0.03 mg/L [27]. The level of phosphate in the lake water

Source: field data, 2017.

may be attributed to high activities such as rapid evaporation and mineralization of decomposed material in the water [23] [28]. More so, the gradual increase in phosphate concentration in the lake water may be also attributed to the use of phosphate fertilizers, detergents and waste from domesticated animals around the lake. This is an indication of anthropogenic pollution [29]. “Reference [30] reported that high values of phosphate support algae growth and may affect the fish in the lake water.”