Annex I
AGENDA

1. Opening of the Session

2. Review of document “Trace Metals in African Aquatic Environments”

3. Feasibility to produce a document on pesticides in the African aquatic environment (herbicides, organo-phosphorous insecticides, artificial pyrethroids, etc.)

4. Role of the Working Party in the CIFA Symposium in Inland Fisheries, Aquaculture and Environment, 1994

5. Any other matters

6. Approval of the Report

7. Closing of the Session

Annex II
LIST OF PARTICIPANTS

Anthony T. AMUZU
Water Resources Research Institute
P.O. Box M.32
Accra
Ghana
Fax (233-21)773068

Charles BINEY
Institute of Aquatic Biology
P.O. Box 38
Achimota
Ghana
Fax (233-21)773068

Davide CALAMARI (Convenor)
Institute of Agricultural Entomology
University of Milan
Via Celoria 2
I-20133 Milan
Italy
Fax (39-2)26680320

Nasséré KABA
Centre de recherches océanologiques
B.P. V18
Abidjan
Côte d'Ivoire
Fax (225)351155

Israël Lape MBOME
Centre de nutrition
Institut de recherche médicale et d'études des plantes médicinales (IMPM)
B.P. 6163
Yaoundé
Cameroun
Telex MESRES 8413KN

Heiner C.F. NAEVE (Technical Secretary)
Fishery Resources and Environment Division
FAO
Via delle Terme di Caracalla
I-00100 Rome
Italy
Fax(39-6)5120330

Peter B.O. OCHUMBA
Kenya Marine and Fisheries Research Institute
P.O. Box 1881
Kisumu
Kenya
Fax (254-11)472215

Oladele OSIBANJO
Department of Chemistry
University of Ibadan
Ibadan
Nigeria
Fax (234-1)6611531

Vivian RADEGONDE
Technological Support Services Division
Department of Industry
Pointe Larue
Mahé
Seychelles
Fax (248)76151

Massoud A.H. SAAD
Department of Oceanography
University of Alexandria
Moharrem Bay
Alexandria
Egypt
Fax (203)815658

W.Q.B. WEST (Secretary CIFA)
FAO Regional Office for Africa (RAFR)
P.O. Box 1628
Accra
Ghana
Fax (233-21)668427

Annex III
LIST OF DOCUMENTS

1. Agenda

2. List of Participants

3. Trace Metals in African Aquatic Environments, drafted by Ch. Biney, Ghana

4. Comments on Draft Document received from A.T. Amuzu, Ghana

5. Comments on Draft Document received from N. Kaba, Côte d'Ivoire

6. Comments on Draft Document received from I.L. Mbome, Cameroon

7. Comments on Draft Document received from O. Osibanjo, Nigeria

8. Comments on Draft Document received from V. Radegonde, Seychelles

9. Comments on Draft Document received from S.O. Wandiga, Kenya

Annex IV
REVIEW OF HEAVY METALS IN THE AFRICAN AQUATIC ENVIRONMENT

by

Ch. Biney, A.T. Amuzu, D. Calamari, N. Kaba, I.L. Mbome, H. Naeve, P.B.O. Ochumba, O. Osibanjo, V. Radegonde and M.A.H. Saad

1. INTRODUCTION

In natural aquatic ecosystems, metals occur in low concentrations, normally at the nanogram to microgram per litre level. In recent times, however, the occurrence of metal contaminants' especially the heavy metals in excess of natural loads, has become a problem of increasing concern. This situation has arisen as a result of the rapid growth of population, increased urbanisation, expansion of industrial activities, exploration and exploitation of natural resources, extension of irrigation and other modern agricultural practices as well as the lack of environmental regulations.

At the global level the scientific community has investigated some of these problems and the results have been published in several reviews and books (Nriagu, 1989; Förstner and Wittmann, 1981; Salomons and Förstner, 1984). Most of the present introductory material is based on these sources.

Unlike other pollutants like petroleum hydrocarbons and litter which may visibly build up in the environment, trace metals may accumulate, unnoticed, to toxic levels. Thus problems associated with trace metal contamination were first highlighted in the industrially advanced countries because of their larger industrial discharges and especially by incidents of mercury and cadmium pollution in Sweden and Japan (Kurland et al., 1960; Nitta, 1972; Goldberg, 1976). In spite of the relatively low level of industrial activity in less developed regions such as Africa, there is nevertheless growing awareness of the need for rational management of aquatic resources including control of waste discharges into the environment. This becomes even more important in view of the expected increases in industrial and urban activities in all parts of the continent.

Existing information on various environmental problems has been reviewed by Dejoux (1988) in a monograph of pollution in African inland waters and by Phillips (1991) on a worldwide basis on tropical marine ecosystems. These publications showed that the existing information on Africa is scattered and scarce, and therefore demonstrated the need for a more precise and specific review of the occurrence of trace heavy metals in various aquatic environmental compartments in the continent.

For effective water pollution control and management there is a need for a clear understanding of the inputs (loads), distribution and fate of contaminants, including trace metals from land-based sources into aquatic ecosystems. In particular, the quantities and qualities need to be considered together with the distribution pathways and fate and the effects on biota.

The need to make an assessment of the level of heavy metal contamination in the African environment has led to the initiation of several pollution monitoring programmes and research work in various universities and scientific institutions in the region. The most relevant programmes are the Mediterranean pollution monitoring programme (MEDPOL) covering also North Africa, the West and Central Africa marine pollution and research programme (WACAF 2) and the Eastern Africa marine pollution and research programme (EAF/6).

During the last decade, there have also been considerable improvements in the sampling and analytical techniques for trace metals. These, coupled with international intercomparison exercises, have facilitated the generation of more reliable data. The present paper thus attempts to compile and analyze the available information on the occurrence of trace metals in both freshwater and marine ecosystems of Africa as a contribution towards the formulation of rational management policies for aquatic resources in the continent.

The decision to review freshwater and marine data jointly is a result of the need to have a holistic approach that could influence future control strategies.

2. SOURCES OF TRACE METALS

Trace metals enter the aquatic environment from both natural and anthropogenic sources. Entry may be as a result of direct discharges into both freshwater and marine ecosystems or through indirect routes such as dry and wet deposition and land runoff. Important natural sources are volcanic activity, continental weathering and forest fires. The contribution from volcanoes may occur as large but sporadic emissions due to explosive volcanic activity or as other low continuous emissions, including geothermal activity and magma degassing (Zoller, 1984). The major sources of atmospheric mercury, for example, are land and ocean degassing (GESAMP, 1988). In view of the toxic nature of the trace heavy metals, the knowledge of their sources and fate in the environment is important.

The anthropogenic sources include:

1. Mining effluents
2. Industrial effluents
3. Domestic effluents and urban storm-water run off
4. Leaching of metals from garbage and solid wastes dump
5. Metal inputs from rural areas, e.g. metals contained in pesticides
6. Atmospheric sources, e.g. burning of fossil fuels, incineration of wastes and industrial emissions
7. Petroleum industry activities.

Table I presents examples of potential industrial and agricultural sources for metals in the environment.

In some African countries, mining activities are important sources of heavy metal input to the environment, for example mercury in Algeria, arsenic in Namibia and South Africa, tin in Nigeria and Zaire, and copper in Zambia.

Table I
Industrial and agricultural source for metals in the environment

For most trace metals, anthropogenic emissions are more than or equal to natural emissions. The combustion of leaded petrol in automobiles, for instance, is responsible for the widespread distribution of lead in the world. For mercury, however, several reports (Hutchinson and Meema, 1987; GESAMP, 1988) suggest that natural emissions are quantitatively more important than anthropogenic sources.

From the above considerations the compilation of an inventory of the sources and the quantification of the heavy metal loads appear to be important tasks to be accomplished. Obviously, this exercise should be carried out on a country-by-country basis, with the ultimate goal of a regional and continental overview of the problem.

Metal concentrations in industrial effluents and landfill leachate in African countries are shown in Table II.

3. DISTRIBUTION PATHWAYS AND FATE OF HEAVY METALS IN THE AQUATIC ENVIRONMENT

Once in the aquatic environment, metals are partitioned among the various aquatic environmental compartments (water, suspended solids, sediments and biota). The metals in the aquatic environment may occur in dissolved, particulate and complexed form.

The main processes governing distribution and partition are dilution, advection, dispersion, sedimentation, and adsorption/desorption. Nonetheless, some chemical processes could also occur. Thus speciation under the various soluble forms is regulated by the instability constants of the various complexes and by the physico-chemical properties of the water (pH, dissolved ions, Eh and temperature).

Adsorption could be the first step in the ultimate removal of metals from water. In the course of distribution, permanent or temporary storage of metals takes place in the sediments of both freshwater and marine environments. Microbial activity and redox processes may change the properties of sediments and affect the composition of interstitial water. As a result, iron and manganese oxides may be converted to carbonates or sulphides, leading to a decrease in the adsorption capacity of the sediments. Reworking of the sediments by organisms will also bring sediments to the surface, where a significant fraction of the metal will be released.

Table II
Mean metal concentrations in some industrial effluents and landfill leachates (μg ml^-1)

Many transformations of heavy metals in aquatic environments occur as biochemically mediated reduction, methylation, demethylation and oxidation of single metal species. Redox reactions may also facilitate some transformations. The biochemical processes are carried out by microorganisms and algae. According to Jernelöv (1975), methylation of mercury takes place when microorganisms, while consuming organic substances, happen to come into contact with mercury ions. This may also be true for As, Sn and Pb.

Heavy metals are taken up by both fauna and flora. This uptake could provoke an increase in the concentration of the metal in the organism; if the excretion phase is slow, this can lead to the bioaccumulation phenomenon. A few metals such as mercury have been shown to undergo biomagnification through the food chain.

4. EFFECTS AND QUALITY CRITERIA FOR METALS

Some heavy metals such as Zn, Cu, Mn and Fe are essential for the growth and well-being of living organisms including man. However, they are likely to show toxic effects when organisms are exposed to levels higher than normally required. Other elements such as Pb, Hg and Cd are not essential for metabolic activities and exhibit toxic properties.

Metal contamination of the aquatic environment may lead to deleterious effects from localised inputs which may be acutely or chronically toxic to aquatic life within the affected area. Most published data on the effects of metals on aquatic organisms, however, report adverse effects at concentrations higher than usually found in the environment (GESAMP, 1985; 1988).

Metals may be taken up in the inorganic or organic form. For some elements, such as arsenic and copper, the inorganic form is the most toxic. For others, such as Hg, Sn and Pb, the organic forms are the most toxic. At low concentrations many heavy metals, including Hg, Cd, Pb, As and Cu, inhibit photosynthesis and phytoplankton growth. Effects at higher trophic levels include delayed embryonic development, malformation and reduced growth of adults of fish, molluscs and crustaceans.

In the last two decades, many investigations have been conducted on the toxicity of metals the results of which have formed the basis for the formulation of water quality criteria for aquatic life by several international and national organizations.

For example, the European Inland Fisheries Advisory Commission (EIFAC) of the Food and Agriculture Organization (FAO) gave the following definition for water quality criteria to protect fisheries:

“Water quality criteria for freshwater fish should ideally permit all stages in the life cycle to be successfully completed, and in addition, should not produce conditions in a river water which would either taint the flesh of the fish or cause them to avoid a stretch of river where they would otherwise be present, or give rise to accumulation of deleterious substances in fish to such a degree that they are potentially harmful when consumed. Indirect factors like those affecting fish-food organisms must also be considered, should they prove to be important”. A number of water quality criteria have been produced by EIFAC (Alabaster and Lloyd, 1982).

The major routes of heavy metal uptake by man are food, water and air. For example, aquatic fauna, especially fish, are the most important source of mercury and arsenic for human beings.

Many publications exist on the effects of heavy metals in humans, and reviews have also been prepared on these subjects by the International Programme on Chemical Safety of the World Health Organization (IPCS/WHO) and disseminated through the Environmental Health Criteria publication series.

The WHO has also produced water quality criteria for drinking water for several chemicals including most of the heavy metals (WHO, 1984) while FAO has made a compilation of legal limits for hazardous substances in fish and fishery products with particular reference to heavy metals (Nauen, 1983).

5. METHODS OF ANALYSIS

Several methods have been used to determine trace elements in environmental matrices. In early studies, gravimetric, volumetric and colorimetric techniques were employed. The more common colorimetric method involved formation of soluble metal complexes, chelates, with such organic compounds as dithizone, o-phenanthroline and ammonium pyrolidine dithiocarbamate (APDC). Modern methods such as anodic stripping voltametry (ASV) and the use of ion-selective electrodes (ISE) are based on electrochemical principles.

Other methods employ nuclear related techniques. These include proton-induced X-ray emission (PIXE), instrumental neutron activation analysis (INAA), X-ray fluorescence (XRF) and inductively coupled plasma mass spectrometry (ICP-MS). Most of these methods are however very expensive, and only a few recent studies in Africa (Kakulu et al., 1987; Akoto Bamford et al., 1990; Onwumere and Oladimeji, 1990) have reported using them.

Also in Africa, by far the most common methods for the determination of heavy metals is atomic absorption spectrophotometry (AAS). It has the advantage of speed, sensitivity, simplicity and ability to analyze complex mixtures without prior separation. For most of the heavy metals the basic technique involves flame atomization, while for some of them at very low concentrations the graphite-furnace technique involving electrothermal atomization is used. Vapour-generation methods may be used for a few metals such as As, Se, and Sb, whereas mercury is analyzed by the cold vapour technique.

6. SELECTED AFRICAN EXPERIENCES

In the following section, selected African experiences have been summarized. These refer to research conducted on distribution of metals in various environmental compartments and illustrate situations in which water bodies are influenced by metal loads.

6.1 Northern Africa

Heavy metals and pesticides in Northern African waters were studied more recently than other chemical parameters. The compartments water and sediments, as well as selected biota of inland water bodies and coastal marine areas, have been investigated. Of the different matrices, sediments have been more analyzed because they can present a clearer indication of metal inputs and accumulation in aquatic environments.

Studies of heavy metals in Northern Africa have been concentrating on Egyptian inland waters and coastal zones, particularly on the river Nile and its two branches (Rosetta and Damietta), as well as on the delta lagoons. However, many studies have been conducted within the framework of the 1975 Action Plan for the protection of the Mediterranean and have therefore focused on the coastal zones. Advanced investigations on the dynamics and speciations of trace metals are also being conducted in different Egyptian inland and coastal marine waters.

Bernhard and Renzoni (1977) differentiated between natural and anthropogenic sources of mercury pollution in the Mediterranean by reviewing concentrations in pelagic fishes and benthic organisms, as well as sediments.

Toma et al. (1980) investigated the distribution of adsorbed metals on the fine fraction of the sediments of the western part of the Nile continental shelf (offshore, near shore and river environments). The authors concluded that abundance of metals occurred in the order Fe > Mn > Zn > Cu, and their distribution was identical with the pattern of sediment transport. High concentrations of metals were recorded at certain localities and some of them, reaching potentially toxic levels for aquatic organisms, were attributed to contaminated drainage waters. The occurrence of some heavy metals in the sediments of Abu-Kir Bay of the Mediterranean Sea was studied by Saad et al. (1981b). The metals (Cu, Cd, Zn, Fe and Mn) showed a pattern of distribution similar to that of the mud and organic matter content of the sediments. The effects of industrial effluents were found to be restricted to sediments in the vicinity of their discharge.

Studies on the surface sediments of El-Mex region of the Mediterranean in front of Alexandria (Saad et al., 1981) revealed two zones, one of which showed high concentrations of Mn, Cu, Cd, Zn and Fe, as a result of discharges of industrial effluents. Their findings also suggested incorporation of similar proportions of Fe and Mn into the sediments and the co-precipitation of Cu and Zn by iron oxides.

The occurrence and distribution of metals in the water of the heavily polluted Lake Mariut in Egypt, and their accumulation in the different parts of a common fish (Tilapia sp.) in this lake were investigated by Saad et al. (1981a). Variations in the concentrations of metals (Zn, Cu, Fe, Mn and Cd) in the lake water were mostly attributed to variations in the discharge rates of the dumped wastes. The levels of these metals in fish were much higher than those in water.

The seasonal distribution of dissolved and particulate heavy metals in the water column of the Damietta branch of the river Nile was studied by Fahmy (1981). El-Rayis and Saad (1985) estimated the contribution of trace metals from the river Nile to the eastern Mediterranean by determining the concentrations of dissolved metals in the surface and subsurface water along the Rosetta branch. The relative abundance was Zn > Fe > Cu > Mn > Cd.

Saad and Fahmy (1985) studied the occurrence of trace metals in surficial sediments from the Damietta estuary of the Nile and concluded that the eastern side of the estuary was exposed to more pollution than the western side. Also, areas of maximum averages of Cu, Zn and Cd coincided with the discharge sites of sewage wastes.

Metal enrichment in surficial sediments of three shallow Nile delta lakes (Lake Mariut, Nozha Hydrodrome and Lake Manzalah) was evaluated using Fe/metal ratios by Saad et al. (1985). A further study on heavy metals in water, sediments and fish of Lake Mariut was made by Saad (1985). The results revealed the existence of a direct relationship between the levels of metals in the lake water and in the different fish parts. The metal concentrations in the lake water were markedly lower than those in fish and the levels, excluding cadmium, in the lake sediments were considerably higher than those in fish. Analyses of sediments, aquatic plants and water from the river Nile and its branches at selected sites characterized by heavy industrialization and dense populations (Fayed and Abd El-Shafy, 1985) showed higher concentration factors in sediments than in plants.

El-Rayis and Saad (1986) studied the levels of heavy metals in a big land-based source of contaminated drainage water contributing six million m³/day to the coastal Mediterranean region in front of Alexandria. Another study by El-Rayis and Saad (1986a) divided Lake Mariut into two zones, septic and non-septic, after metal analysis of suspended matter and water. According to El-Rayis et al. (1986), copper and zinc were found to be concentrated in the sandy sediments of the shallow sides of the Eastern Harbour of Alexandria, whereas iron and manganese occurred in the deeper sediments.

Saad (1987) studied heavy metals in the Nozha Hydrodrome by analyzing water, sediments and different parts of Tilapia sp. to gain information on the levels of these metals and their possible variations in relation to pollution effects from discharges of polluted Nile water into the lake. The results suggested a direct relationship between the concentrations in water and in fish. They also illustrated the ability of fish to absorb high levels of heavy metals and the ability of lake sediments to accumulate these metals. El Rafei et al. (1987) quantified the levels of trace metals in waste waters discharged into the river Nile from some industries near Cairo. El-Nabawi et al. (1987) determined metal concentrations in fish from Lake Mariut, Lake Edku and Abu-Kir Bay and found the highest levels in Sphyraena sphyraena from this bay. Moharram et al. (1987) estimated the levels of total, organic, inorganic mercury, total selenium and the interaction between both metals in Mugil cephalus. A strong correlation was reported between fish length and each of these variables.

Data on trace metals in the Red Sea are scanty. Saad and Kandeel (1988) investigated the distribution of Cu, Fe and Mn in the coastal Red Sea waters in front of Al-Ghardaqa. The different patterns of seasonal distribution and the irregular variations of these metals were discussed.

Heavy metal pollution in Lake Mariut has been further investigated by El-Rayis and Saad (1990), based on the distribution of Cu, Zn, Fe and Mn in water, suspended matter and sediments. The contribution of metals from this lagoon to the Mediterranean Sea via Umum Drain (contaminated land-based source) was also estimated.

6.2 West and Central Africa

Studies on the occurrence and distribution of metals in Nigeria have been conducted on all the major environmental matrices (water, sediment, fauna and flora) but again with more emphasis on sediments.

Statistical treatment of the result of metal analyses of 176 stream sediment samples from the lfe-llesha area (1800 km²) of southern Nigeria (Ajayi, 1981) showed that all the elements have density distribution close to natural background levels. Ojo (1988) also used various statistical methods for the interpretation of the geochemical data obtained from analyses of Cu, Pb, Zn, Co, Ni, Fe, Mg, Mn and Ca in 374 stream sediment samples collected over an area of 700 km² within the upper Benue Trough (Nigeria) and concluded that these elements exhibit various patterns of association depending on their nature and prevailing environmental conditions. Other studies in the area (Kakulu and Osibanjo, 1988, 1991) revealed elevated levels of Pb, Cr, Ni, V and Zn in Port Harcourt and Warri sediments which suggest that effluents from petroleum refineries located in these cities have contributed significantly to the heavy metal pollution of the respective aquatic ecosystems.

Okoye et al. (1991) reported anthropogenic heavy metal enrichment of Cd, Co, Cu, Cr, Fe, Mn, Ni, Pb and Zn in the Lagos lagoon and implicated land based urban and industrial wastes sources.

Pollution studies on 26 rivers in some southern and northern states of Nigeria (Ajayi and Osibanjo, 1981), on rivers in the Niger Delta (Kakulu and Osibanjo, 1991), on the cocoa growing area of Ondo State in South West Nigeria (Ogunlowo, 1991) and the Lagos waters (Okoye, 1991a) showed that, with the exception of iron, the concentrations of most trace metals in the surface waters are generally lower than the global average levels for surface waters and the international drinking water standards.

Ndiokwere and Guinn (1982) determined As, Cd, Cr, Hg, Mn, Mo, Ni, Se and Sb in two Nigerian rivers and two harbours and attributed high metal concentrations to local pollution sources. In their studies of streams and lakes around ibadan, Mombeshora et al. (1983) reported much higher levels of lead in sediments than in water. The highest levels of lead coincided with areas of high traffic density.

Analyses of sediments and fish from the Niger delta area of Nigeria (Kakulu and Osibanjo, 1986) revealed that the area was relatively unpolluted with mercury compared to some European areas (Mediterranean, Baltic Sea and North-East Atlantic). Report from the same area (Kakulu et al., 1987a) indicated that the levels of Cd, Cu, Fe, Mn, Pb, and Zn were higher in shellfish than in finfish. With the exception of the lead levels in some shellfish, levels of these metals were generally lower than the WHO recommended limits in foods. Concern about the high levels of lead in Lagos lagoon fish has also been expressed (Okoye, 1991).

Other Nigerian studies include that of Sridhar (1988) who analyzed the aquatic plant Pistia stratiotes and showed that the shoot system accumulated more K, Ca, and Mg, whereas the root system accumulated significantly more Cd, Cr, Co, Fe, Pb, Hg, Na, and Zn.

Onwumere and Oladimeji (1990) reported that Oreochromis nilotica exposed to treated petroleum refinery effluents accumulated trace metals in the order Pb > Cu > Zn > Mn > Cr > Ni > Cd.

In other parts of West Africa, the concentrations of major and minor ions, including Cu, Mn and Fe in river Jong, Sierra Leone, was determined by Wright (1982), who found a clear relationship between metal concentrations and seasonal variations in rainfall.

In Ghana, one of the earliest studies (Amasa, 1975) examined various matrices, including drinking water, from the Obuasi gold mining area and found that arsenic concentrations occurred above normal values. A more recent study (Akoto Bamford et al., 1990) in which heavy metal pollution from gold mining activities was assessed by analyzing gold ore, tailings, sediments and water for Cr, Mn, Fe, Cu, Zn, As, Pb, Rb, Sr, Y, Zr and Nb, revealed the presence of all the elements in sediments within a concentration range of 0.08 to 49 000 μg g^-1, whereas only iron and zinc were detected in water at levels of 0.08–2.4 (μg ml^-1).

Total mercury concentrations in commercial fish from different coastal sites of Ghana have been determined by Ntow and Khwaja (1989) who concluded that all values were well below the 0.5 μg g^-1 action level adopted in many countries. Biney and Beeko (1991) conducted a survey of metals in fish and sediments from the River Wiwi in Kumasi and found a positive correlation between mercury concentration and body weight of fish. They also reported higher levels of cadmium and mercury in fish than in sediment. Studies on the distribution of Hg, Cd, Pb, Cu, Zn and Fe in water, finfish and shellfish, macrophytes and sediments from Kpong headpond and lower Volta river (Biney, 1991) showed the highest concentration of iron and lead in sediments and of manganese and cadmium in macrophytes. Finfish had the lowest concentrations of the metals, except for lead.

Pelig Ba et al. (1991) assessed the level of contamination of drinkable ground-water from the Accra plains and upper regions of Ghana and found that in some areas Pb, Cr and Fe concentrations exceeded the WHO guideline limits for drinking water.

In Cote d'Ivoire Marchand and Martin (1985) and Kouadio and Trefry (1987) have studied sediments of the Ebrié Lagoon and reported metal concentrations in excess of background levels, this was attributed to the disposal of untreated sewage and industrial effluents.

A comparative study by Metongo (1991) of Cd, Cu, Hg and Zn in samples of oysters (Crassostrea gasar) from urban and rural lagoon areas of Côte d'Ivoire revealed higher but background levels of the metals in the urban area. Likewise, other studies of heavy metals in Callinectes amnicola (Metongo and Sankaré, 1990) and in Thunnus albacares (Metongo and Kouamenan, 1991) gave concentrations lower than internationally acceptable limits for seafood.

In Senegal, analyses by Gras and Mondain (1978) of fish and crustaceans from coastal waters revealed lower mercury concentrations than the generally acceptable limits (0.5 μg g^-1), except in swordfish and sharks weighing more than 5 kg.

Other studies on the occurrence of trace metals have been conducted as part of the Joint FAO/IOC/WHO/IAEA/UNEP Project on monitoring of pollution in the marine environment of the West and Central African region. Within this framework, concentrations in marine biota have been reported for Cameroon (Mbome et al., 1985; Mbome, 1988), Ghana (Biney, 1985; Biney and Ameyibor, 1989) Côte d'Ivoire (Metongo, 1985, 1988) and Senegal (Ba et al., 1985; Ba, 1988). On the basis of these studies, Portmann et al. (1989) reviewed the levels of contaminants in the marine environment of the region and concluded that there was little input of mercury and other metals into the coastal zone from land.

6.3 Eastern Africa

Early studies in this region focused on Lake Nakuru in Kenya, one of a number of soda lakes in the Great Rift Valley which was made a national park in 1986 because of its world-famous flamingo population.

In an attempt to produce baseline information for monitoring pollution, Koeman et al. (1972) determined As, Sb, Cu, Zn, Cd and Hg in muscle, liver and kidney of birds and fish. They concluded that the metal concentrations did not constitute a hazard to the biota of Lake Nakuru. Six years later, Greichus et al. (1978a) studied water sediment, benthos and fish, and reported slightly elevated concentrations as compared to values found by Koeman et al. (1972).

The effect of copper ions on the photosynthetic oxygen production of phytoplankton, on the growth rate of blue-green algae (Spirulina platensis) and on populations of rotifers (Brachionus sp.) in water from Lake Nakuru was experimentally investigated by Kallqvist and Meadows (1978). The rotifers were less sensitive to copper than algae. Other studies by Lewin (1976) showed that Lake Nakuru water contained 0.08 mg Cu l^-1 mainly from pesticide containing run-off from the surrounding agricultural lands. This value was thus higher than the critical value of 0.02 mg Cu l^-1 which may significantly reduce algal growth (Kallqvist and Meadows, 1978). Observations by Ochumba (pers.comm.) have shown that, during the dry season, flamingos which feed on the algae migrate away from the lake. This may negatively affect the tourism industry.

Earlier studies on sediment, water and biota of the second largest natural lake in the world, lake Victoria (Alala, 1981; Onyari, 1985; Ochieng, 1987) showed no significant heavy metal pollution. However, more recent studies in the same area revealed increased lead levels largely due to increased shipping traffic and associated problems, car washing and discharge from local industries (Wandiga and Onyari, 1987; Onyari and Wandiga, 1989). Ochumba (1987) studied physico-chemical parameters, dissolved oxygen and heavy metal concentrations in Lake Victoria as the possible causes of periodic fish kills. The author attributed the fish kills to dissolved oxygen depletion.

In other East African areas, copper ion distribution in the surface waters of Lakes George and Edward (ldi Amin) in Uganda was studied alongside other chemo-limnological parameters (Bugenyi, 1979). Concentrations ranged from 0.07 to 0.13 μg ml^-1 in Lake George and from 0.006 to 0.02 μg ml^-1 in Lake Edward. A direct relationship was established between copper, water hardness, alkalinity and total dissolved solids. Bugenyi (1982) studied the occurrence of Cd, Cu, and Fe in sediments of the same lakes and concluded that the concentrations, although distinct in the different water bodies, did not show much variations within each of the lakes.

Effluent, air and soil samples near a battery factory in Dar-es-Salaam, Tanzania, were analyzed for mercury by Semu et al. (1986). The highest levels of contamination were associated with the disposal of defective batteries. A preliminary investigation of the extent of metal pollution of the Msimbazi river in Dar-es-Salaam, which receives industrial, urban and agricultural waste waters, was conducted by analyzing sediments and biological indicators (Akhabuhaya and Lodenius, 1988). Metal concentrations were in general low but some of the results indicated localised industrial pollution.

Studies of dissolved metals in the marine environment were conducted by Norconsult (1977) concluding that the concentrations for Tudor Creek fell within the normal range of unpolluted natural sea water. Oteko (1987) studied the Mombasa Creek and suggested crustal sources to be responsible for copper concentrations and increased anthropogenic sources from automobile exhausts for cadmium and lead concentrations.

According to Bryceson et al. (1990) available data on marine contaminants are scarce. However, localised hot spots of metal pollution are found in the vicinity of cities and industrial centres that may constitute a danger to the public health. Wandiga and Onyari (1987) found slightly higher metal concentrations in marine fishes from Mombasa when compared to fish from Lake Victoria. The reported concentrations did not pose an immediate danger to the fish industry.

Matthews (1981) found evidence of surprisingly high mercury levels in fish (1.0 – 2.0 μg g^-1) and in hair and blood of inhabitants of Seychelles, where fish consumption is very high. The sources and pathways of such high mercury levels are a mystery.

6.4 Southern Africa

The concentrations and distributions of metals amongst other chemical contaminants were investigated by Greichus et al. (1977) in two South African lakes; Hartbeespoort dam, which receives industrial and municipal waters from Johannesburg and Voelvlei dam, situated in mainly agricultural area. Water, sediment, aquatic plants and insects, fish, fish-eating birds and their eggs were analyzed for As, Cd, Cu, Mn, Pb, Zn and Hg. The results indicated higher levels in Hartbeespoort dam than in Voelvlei for all metals in sediments and birds, except for copper in bird carcasses. Mercury levels in birds were 2 to 5-fold greater than in fish, whereas lead values were 2 to 10-fold greater.

Greichus et al. (1978) investigated metals among other contaminants in Lake Mcllwaine, a eutrophic water body near Harare, Zimbabwe. Water, sediment, plankton, bottom fauna and fish were analyzed. The data gave intermediate levels of metals between those found in Hartbeespoort dam and Voelvlei dam.

Watling and Emmerson (1981) identified areas of metal input to the river Papenkuils which was considered to be a serious source of pollution to the marine environment around Port Elizabeth. In contrast, the estuary of river Swartkops was found generally unpolluted on the basis of metal concentration in water, surface sediments and sediment cores (Watling and Watling, 1982). Similar studies also showed that the estuary of river Knysna as well as the Bushmans, Kariega, Kowie and Greatfish rivers were unpolluted (Watling and Watling, 1982a, 1983).

7. LEVELS OF HEAVY METALS IN DIFFERENT ENVIRONMENTAL COMPARTMENTS

In chapter 6 a qualitative presentation of selected African experiences by regions has been outlined; in chapter 7 a more detailed discussion will be presented on the quantitative aspects based on concentration levels in various environmental compartments.

In general, studies on the levels and distribution of contaminants including heavy metals in Africa have been concentrated in urban and industrial areas. Thus the actual background levels in water, sediments and biota may still not be accurately known and this fact can also produce bias in the interpretation of the data.

7.1 Concentrations of metals in water

Data of dissolved metals in inland water bodies (lakes and rivers) and coastal marine zones are presented in Table III. More data are available for inland than for coastal waters. As already pointed out, most studies on the levels and distribution of heavy metals in Africa have concentrated on urban and industrialized areas.

Mercury showed the lowest concentrations <1.0μg l^-1), followed by cadmium (0.2–21.0 μg l^-1). Iron gave the highest levels in most waters with a considerable wide range of variation, from 2.5 μg l^-1 in the river Nile, Egypt (El-Rayis and Saad, 1985), to 14,400 μg l^-1 in Shasha stream, Nigeria (Martins, 1978). The other metals are generally arranged in the following order of abundance: Mn > Zn > Pb > Cu > As. The levels of metals in the coastal waters were markedly lower than those found in most inland waters. This reflects the direct influence of pollution on the lakes and rivers. Undoubtedly, the very high concentrations of certain metals found in specific waters are a result of acute pollution.

7.2 Concentrations of metals in sediments

On a continental basis, heavy metal inputs into African inland water sediments (Table IV) pose no great concern in relation to unpolluted sediments (GESAMP, 1982).

Most water bodies showed low-to-moderate metal concentrations except for Hartbeesport dam and river Papenkuils in South Africa, the Niger delta in Nigeria and lakes George and Edward (Idi Amin) in Uganda. Lakes Nozha, Mariut and Manzalah and the river Nile in Egypt as well as lake Mcllwaine in Zimbabwe also showed elevated concentrations of some metals which clearly indicate considerable anthropogenic inputs.

Metal concentrations in inland water sediments thus pose no environmental concern for the continent except for the above-mentioned areas which may be considered as hot spots within their respective regions.

Table III
Mean dissolved metal concentrations in inland and coastal waters (ng ml^-1)

* Range values

Table IV
Mean metal concentrations in inland water sediments (μg g^-1 dry weight)

* Range values

Heavy metal concentrations in the African marine and coastal sediments (Table V) fell within the ranges given for Hg, Cd, Pb and As by GESAMP (1985, 1988) but higher values occurred in some areas. For example, sediments from Lagos lagoon in Nigeria had high concentrations of lead and iron while the Ebrié Lagoon in Côte d'Ivoire had high mercury, zinc and iron concentrations. The results revealed largely anthropogenic heavy metal enrichment implicating urban and industrial runoff into coastal lagoons which have poor water exchange (Okoye, 1989; Kouadio and Trefry, 1987). However, data for Egypt (Saad et al., 1981, 1981b) and the Nigerian Atlantic coast (Ndiokwere, 1984) did not suggest high contamination of the coastal sediments.