Persistent organic pollutants in water, sediments and fish from Lake Victoria and implications to human health risks

Thesis 2019

Lake Victoria is an important freshwater resource in Tanzania and its neighbouring countries namely Uganda, Kenya, Rwanda, Burundi and Democratic Republic of Congo (DRC). Millions of people dotted around the lake depend on it as a major source of household and industrial water supply, irrigation, transport and provides for fishing as a main source of animal derived protein as well as income. In Tanzania, the lake provides employment opportunities to more than four million people engaged directly or indirect with fishing activities. The frequently fished and consumed fish species that are of commercial interest include; Nile perch (Lates niloticus), Nile tilapia (Oreochromis niloticus) and the freshwater sardines (Rastrioneobola argentea) commonly known as dagaa in Tanzania. The fish are further processed to feed the domestic, regional and international markets. Current estimates from the lake put the annual contribution of the fisheries sector to the Tanzanian Gross Domestic Product (GDP) up to 2.5%. There have been unsubstantiated claims that despite using agrochemicals such as pesticides in agriculture, some fisher folks have been using such chemicals for fishing and preservation of fish products during storage. The same organochlorine pesticides are still in use in agricultural activities within the lake basin and there is a possibility that they accumulate in water, sediments and fish. Furthermore, there have been incinerations of medical wastes, discharge of electronic wastes and untreated wastewaters from industrial establishments in water bodies that tend to increase the loading of polychlorinated biphenyls (PCBs) in water, sediments and fish. It is likely that human beings will get exposed to these chemicals through drinking water and eating fish from the lake. The present study overall objective was therefore to assess fish consumersꞌ eating habits and the prevalence, levels and human health risks of indicator PCBs and OCPs in water, sediments, fish and fish products from Lake Victoria in Tanzania. The specific objectives of the study were: i. To establish levels of indicator PCBs and OCPs in water, sediments and fish from Lake Victoria. ii. To determine the prevalence and quantify residual levels of indicator PCBs and OCPs in processed fish products from Lake Victoria. iii. To assess fish consumersꞌ eating habits. iv. To evaluate the probable human health risks associated with consumption of persistent organic pollutants (POPs) contaminated fish and fish products from Lake Victoria. Water and sediment samples were collected from twelve nationally designated sampling points (considered as important hotspots). Thirty six samples of both water and sediments (a total of 72 samples) were collected for analysis. A total of 162 fresh fish samples were collected from nine selected landing sites and 480 samples of processed fish products from Kirumba Fish Market for analysis of POPs. Extraction of the samples was done using a QuEChERS method and POPs analysis using a gas chromatography equipped with electron capture detectors (GC-ECDs) and a gas chromatography equipped with mass spectrometry (GC-MS). A questionnaire was also administered to 122 fish consumers to assess the fish consumption habits and patterns. The main focus of administering the questionnaire was to identify the types of fish species consumed, frequencies of fish consumption, quantities of fish consumed and factors affecting fish consumption in the population around the lake. The results showed that some organochlorine compounds were present in water, sediments, fish and fish products from Lake Victoria in Tanzania. This is likely due to human anthropogenic activities being undertaken along the lake, long range atmospheric transport and environmental persistence of the compounds. For water samples, the ∑PCBs ranged between 0.95 and 2.24 μg/L while for sediments the range was between <LOD and 10.28 μg/kg- dry weight (dw) basis being dominated by CB 138 and CB 153 because of their structures and high degree of chlorination. For OCPs; Aldrin and Dieldrin were the highest (6.13 ±0.90 and 6.82±0.53 μg/kg- dw, respectively) and p, pꞌ- DDT was the lowest (0.82±0.12 μg/kg- dw) in sediments whereas for water samples the highest concentration was γ- HCH (1.60±0.32 μg/L) and the lowest was α- endosulfan (0.15±0.01 μg/L). There were high levels of organochlorine compounds in sediments than the water samples suggesting that sediment serves as a sink for organochlorine pollutants and therefore act as a source of these compounds in case of environmental changes. However, the levels of POPs measured in water and sediments in this study were far below the Maximum Residual Limit (MRL) set by FAO/WHO and the European Union (EU) except Aldrin and Dieldrin through which based on the Threshold Effect Concentration (TEC) for freshwater ecosystems, Aldrin and Dieldrin are the only OCPs that seem to be a threat to Lake Victoria environment. For the fresh fish species which were considered in the current study; L. niloticus and O. niloticus, the results revealed that fish species sampled were undersize, reflecting overfishing and abuse of bylaws. The abuse may be going beyond overfishing and touching other environmental protection bylaws. The results showed further that four organochlorine pesticides (β-HCH, HCB, Aldrin and Dieldrin) were identified in fresh muscles of L. niloticus and O. niloticus at measurable quantities at concentration between <0.24 and 1.19 μg/kg. There were slightly higher levels of the identified OCPs in L. niloticus than in O. niloticus muscles due to differences in trophic levels and feeding habits. However, the levels of OCPs measured were below the MRL of 200 μg/kg set for fish by FAO/WHO and other competent authorities suggesting that the fresh fish muscles were safe for human consumption. The seven indicator PCBs considered in this study were not detected in all fresh fish species investigated. Low levels of the detected residues and non detection of many POPs which were considered indicates that currently POPs pollution in Lake Victoria has not reached alarming concentrations but requires a strict continuous monitoring to minimize contaminants loading. Nine OCPs were detected at different measurable concentrations in different processed fish products available in the markets from Lake Victoria. The highest amount was p, pꞌ- DDE (3.73±0.31 μg/kg) in deep fried O. niloticus while the lowest level was α- HCH (0.14±0.06 μg/kg) in smoked O. niloticus. The OCPs were more prevalent in smoked products compared to other processed fish products. This is attributed to a reduced co- distillation of the compounds with water vapour during smoking. However, there was high prevalence of some OCPs in deep fried products compared to other fish products. This is probably due to illegal use of some of these chemicals for fish preservation during storage. In this study, the levels of OCPs in processed fish products were below MRL set by different international statutory bodies for fish and fish products implying that the fish products are safe for human consumption with reference to OCPs concentrations. For the seven indicator PCBs which were considered in this study, six of them were detected at different levels in processed fish products. The total PCBs loading ranged from 16.05±3.04 μg/kg (deep fried products) to 31.55±16.66 μg/kg (trims). There was a significant difference (p<0.05) in PCBs loading for deep fried products (with significantly lower levels) compared to other products which were investigated in this study. This observation is mainly because at high temperatures, cooking oil serves as an extracting solvent and therefore the PCBs are transferred the cooking oil. The PCBs loading was dominated by CB 138, CB 153 and CB 180. The same PCBs were more prevalent (between 20% and 80%) in all fish products than other congeners. This is mainly due to their structures, chemical stability and high degrees of chlorination. However, the mean levels of ∑PCBs in this study were below the MRL of 75 μg/kg set for fish and other fish products by the European Commission indicating that the products are safe for human consumption with respect to indicator PCBs. Lifetime cancer risks for both OCPs and PCBs in adults and children were low while the non cancer risks associated with consumption of contaminated fish and fish products were insignificant as the hazard indices (HI) were very low (<1). The risk is considered significant if HI is greater than 1. Results on fish consumer’s preferences, quantities and types of fish consumed and factors affecting fish eating habits in this study showed that L. niloticus, O. niloticus and R. argentea are widely consumed fish species. The average per capita fish consumption pervii day was quantified to be 0.37 kg (~135 kg/year). This level is far greater than the national, worldwide and the per capita fish consumption of a great fish consumer reported in literatures. The main reasons for this high fish consumption in the study area were found to be easy accessibility and availability of different fish species. The major factors affecting fish consumption habits were reported to be price (affordability), convenience, accessibility, availability, nutritional and healthy concerns. The study revealed a slightly higher proportion of male fish consumers (54.1%) than female fish consumers (45.9%) implying that males usually eat outside their homes better meals than home meals. The study revealed further a direct relationship between fish consumption and age of consumers due to knowledge and health reasons and an inverse relationship between fish consumption and family size due to economic reasons. In order to determine the actual human health risks associated with consumption of fish products from Lake Victoria for consumers dotted around the lake, it is therefore necessary to employ the actual per capita fish consumption of 0.37 kg established herein for risk computation rather than the hypothetical per capita fish consumption rate of 0.02 kg reported in literature. Data generated from this study will therefore help decision makers to devise more appropriate and cost effective risk mitigation measures that will minimize the build-up of these organochlorine compounds in the aquatic ecosystems to safeguard consumer's health and the environment. The results will also help to create awareness of the actual adverse effects associated with the established organochlorine compounds among stakeholders in the fisheries sector.