Pesticide Pollution Remains Severe after Cleanup of a Stockpile of Obsolete Pesticides at Vikuge, Tanzania
Report Sara Elfvendahl, Matobola Mihale, Michael A. Kishimba and Henrik Kylin
High levels of DDT residues and hexachlorocyclohexanes (HCHs) were found in soil, well water, and surface wa-ter around a collapsed pesticide storage shed at Vikuge Farm, Tanzania. Residues of DDT and HCHs were found at three soil depths down to 50 cm. Surface soil samples contained up to 28% total DDT and 6% total HCH resi-dues. Water samples had concentrations of up to 30 µg L-1 of organochlorine pesticides. Other compounds de-tected were aldrin, azinphos-methyl, carbosulfan, γ-chlor-dane, chlorprofam, heptachlor, hexazinone, metamitron, metazachlor, pendimethalin, and thiabendazole. Although the visible remains of pesticides have been removed, the remaining soil is itself hazardous waste and poses a risk to the environment and the inhabitants of the surrounding vil-lages. These findings show the necessity to follow up the environmental situation at former storage sites of obsolete stocks of pesticides, and that the environmental problems are not necessarily solved by removing the visible remains.
INTRODUCTIONAfter the development of modern pesticides at the time of World War II, worldwide use has increased in response to a fast growing population, increased demands on food produc-tion, lowering of food production costs, and for vector and lo-cust control. There are, however, several risks associated with increased pesticide use. Obvious problems are risks for envi-ronmental contamination and development of pesticide resis-tant populations of pests. Of particular relevance to developing countries are the risks for increased numbers of inadvertent poisonings due to a low level of awareness of safe pesticide handling and protection. Also, accumulated stockpiles of ob-solete pesticides are a substantial problem in many developing countries and countries in economical transition. This is re-garded as one of the major environmental hazards and a severe threat to human health in these countries (1). Large stocks of obsolete pesticides have accumulated over the last few decades and are often referred to as a ‘historical burden’ for these countries (2). Obsolete pesticides are stocked pesticides that can no longer be used for their intended pur-pose; they are unwanted and have expired and therefore require disposal (3). Many of these stocks have seriously deteriorated, are stored under inadequate conditions, are currently a source of severe pollution, and are a threat to human health and the environment. Often the products have been prohibited or re-stricted for health or environmental reasons, e.g. after banning, withdrawal of registration, or policy decisions by stakeholders. Products may also have deteriorated as a result of improper or prolonged storage. Consequently, the products can neither be used according to their label specifications, nor can they easily be reformulated (3).
In many cases, the pesticides have originated from foreign aid programs of international organizations as donations for preparedness for, e.g. locust or vector control measures and for agricultural use. In the 1960s and 1970s, some tropical coun-tries received large donations of DDT and malathion for ma-laria control programs (4). In some cases there were even ex-cessive donations without examining the actual need for these products in the recipient country. Sometimes donations were done with no or little prior arrangements for distribution and storage of the pesticides. In the worst cases, they were ‘smart’ donations of surplus pesticides aimed primarily at solving en-vironmental problems in the donor country with less regard to the needs of the recipient (3).
Several international organizations, e.g. the Food and Ag-riculture Organization of the United Nations (FAO), United Nation’s Environment Program (UNEP), Organization for Economic Co-operation and Development (OECD) Working Group on Pesticides, have lately paid much attention to the problems with obsolete pesticides. FAO has, with help from local authorities in the countries in question, conducted sur-veys to map the unwanted and expired stocks of pesticides in developing countries. After an inventory in the early 1990s, FAO estimated that more than 100 000 tonnes of obsolete pesticides are stocked in non-OECD countries (1). Current estimates suggest that there are up to 500 000 tonnes of ob-solete pesticides in these countries (3). Of these, at least 40 000 tonnes are found in Africa, but many stocks are yet to be located and examined. In many of the cases described the situ-ation is very serious with great risks to human health and the environment. In many countries, a considerable part of the total obsolete pesticide stockpiles are made up of persistent and bioaccumu-lating organochlorine compounds such as DDT. They represent a problem because of their persistence in the environment and their potential for accumulating in lipid tissues of biota. They may also undergo biomagnification, i.e. the concentrations of the pollutant may rise in organisms at higher trophic levels (5), and they can be transported globally in the atmosphere (6). Although the persistent organochlorine pesticides have re-ceived much international attention there are problems with other types of pesticides as well. But the risk posed by other pesticide classes present in the stockpiles is often more due to their acute toxicity than their persistence. In Tanzania, there are at least 1100 tonnes of pesticide waste, 20% of which is organochlorines (3). In 1998, a national inventory of chemical waste and storage by the National Envi-ronmental Management Council (NEMC) of Tanzania identi-fied the stock of obsolete pesticides at Vikuge State Farm as one of the most problematic cases (7, 8). A soil sample taken at 1 m depth, contained over 100 mg kg-1 of p,p’-DDT. The situation originated in 1986 when partly expired pesticides
were donated from Greece to Tanzania and placed in a shed at Vikuge. Much of the DDT was delivered in small household size packages with Greek labeling, impossible to read for those handling the material in Tanzania. Some packaging was empty already upon arrival. The observed situation at Vikuge in 1989 and 2001 showed that most of the donated pesticides were left unused in an open shed, which eventually collapsed. In 1995, a bush fire partly burned the pesticides, and in 1996 a new better storage was built by NEMC with help from the Swedish In-ternational Development Cooperation Agency. The remaining pesticides were collected as carefully as possible and repacked and placed in the new building. However, the soil around the old storage shed still lacks vegetation and has a strong smell and scattered fragments of pesticide packaging and dead in-sects. Prior to the donation Vikuge State Farm produced seeds, although at the time of the donation there was no activity on the farm. Today, the farm produces hay, of which most is sold as cattle feed in Dar es Salaam. There is a risk of human exposure to the pesticides through drinking water from wells close to the farm. Drinking water is also provided by a distribution center to a cistern at the farm three times a week. The distribution of water from the cistern to the village is via a PVC pipe that runs in a shallow ditch that also drains the most severely contaminated part of the old stor-age area. The small, mainly temporary streams in the area are tributaries to the Ruvu River, which discharges into the Indian Ocean (Fig. 1). It is necessary to gain knowledge on environmental prob-lems concerning sites where obsolete pesticides are stored and how these can be handled to reduce the risks in a tropical cli-mate for the local environment and inhabitants. Not much is known about the type and quantity of pesticides remaining at Vikuge, although DDT, telodrin, and organophosphorous pesticide containers among others have been observed. Infor-
mation about the pesticides sent to Vikuge was archived at the Ministry of Agriculture in Tanzania. Unfortunately, this information has not been possible to locate for the purposes of this study.
The situation at Vikuge State Farm in 1989. A) The storage shed. B) Ex-amples of pesticide containers. Note DDT-container with Greek la-beling, Telodrin and Gusathion (azinphos-methyl). C) Empty con-tainers from the original consign-ment. Photos B. Paulsson. D) The new storage shed May 2001. Photo: H. Kylin.
Figure 1. Location of the contaminated area in comparison to Vikuge village and local water resources. Numbers 1 and 2 show loca-tions of the wells. Yellow areas represent arable fields.
Vikuge State Farm is located in the Kibaha District of the low-land Coast Region, approximately 60 km northwest of Dar es Salaam (6° 47’ S, 38° 52’ E) (Fig. 1). The altitude varies from 100 to 140 m a.s.l. and the contaminated site is on a small hill. Sampling was carried out in August 2000 and in March and May 2001. Soil samples were taken at three depths (0–5 cm, 20–25 cm and 50–55 cm) at five locations within the old stor-age site. Additional samples at the same depths were taken 20 m and 50 m away from the old storage site. Samples of deep soil were taken by digging a hole down to below the deepest sam-pling layer. Samples were then taken with a small, clean spade horizontally inserted from the hole (9). Also, eight surface soil samples (0–5 cm) were taken along a shallow ditch draining the storage area in which the PVC-pipe distributing drinking water from the cistern to the village runs. The samples were wrapped in aluminum foil, put in plastic bags, and refrigerated or fro-zen if extraction was not performed immediately after sampling. Soil properties were determined on a sample taken about 100 m away from the old storage site where the soil was assumed to be similar but less contaminated. The soil at Vikuge is a sandy loam (84.5% sand, 2.7% silt, 11.3% clay, 0.4% organic matter) according to the FAO system (10). Total carbon was 0.24% and organic carbon 0.22%. The soil pH was 3.9 measured in 1 M potassium chloride and 4.5 in deionized water. Water samples were taken from a pond (surface water) near the old storage site and from two dug wells from which local in-habitants take water for cooking and drinking. Well 1 is located near the storage site, and Well 2 close by a rice field somewhat further away. Water samples from the cistern were taken from a tap in Vikuge village, i.e. after it had passed via the PVC-pipe through the contaminated area. Water samples were taken in 1-L glass bottles with Teflon lined screw caps and were preserved with sodium chloride (100 g).
Chemicals
All solvents used were of pesticide or analytical grade (Merck Eurolabs, Spånga, Sweden). Ammonium chloride, sodium chlo-ride, sulfuric acid and anhydrous sodium sulfate (dried at 400°C overnight) were all analytical grade (Merck Eurolabs). The
screening included a total of 87 different pesticides or pesticide metabolites (11) and authentic standards were from Dr Ehren-storfer (Augsburg, Germany).
Analysis
Soil samples (20 g wet weight) were soaked for 15 min in aqueous ammonium chloride (14 ml, 0.2 M) to open up the soil structure. The mixture was shaken vigorously with cyclohexane:acetone (1:1, 100 ml) in a stoppered glass flask. The flask was shaken vigorously for one minute every 10 minutes for 1 hr and then sonicated for 10 minutes in an ultrasonic bath after which the soil was left to settle and the aqueous and organic phases to separate. Unfiltered water samples (1 L) with added sodium chloride (100 g) were extracted three times with ethyl acetate (120 + 60 + 60 ml). In both cases the organic phase was removed and dried using anhydrous sodium sulfate after which the solvent was reduced on a rotary evaporator (9). The final extracts of water were adjusted to 2 ml in cyclohexane:acetone (9:1), while soil extracts, due to the very high concentrations of organochlorine pesticides, were diluted to concentrations equal to 10-3 – 10-7 g soil ml-1. Solvent and matrix blanks were run for all sample types (9). Recoveries of HCHs (11) and DDTs (11) from spiked samples were tested on a reference soil from Mbezi Luis (Tanzania) and deionized water. Recoveries were for the HCHs 60–80% from soil and 80–105% from water, and for the DDTs 70–100% from soil and 95–105% from water. All samples were analyzed using a Hewlett Packard 5890A gas chromatograph equipped with two 63Ni electron-capture de-tectors (GC-ECD) and a Varian 3400 gas chromatograph with two thermoionic (nitrogen-phosphorus) detectors (GC-NPD). Chromatography was performed simultaneously on two capil-lary columns (CP-Sil 19 CB and CP-Sil 5 CB, 0.25 mm x 30 m x 0.25 µm, Chrompack, Nacka, Sweden) attached to the same injector. Injection was splitless and injector and detector tem-peratures were 250°C and 300°C respectively. The temperature program was 90°C for one minute, 30°C min-1 to 180°C, 4°C min-1 to 260°C, isothermal for 12 min.
Limits of Detection and Quantification
Because of the high, but varying levels, particularly of DDTs, but also HCHs and occasionally other pesticides, general de-
Table 1. Concentrations (mg kg–1 dry weight) of pesticides in soil from Vikuge State Farm, sampled in August 2000. Sampling locations A - E are located within the old storage site, and locations F and G are located 20 and 50 m, respectively, away from the old storage site. Figures given as less than (<) indicate presence of the compound, but below the limit of quantification for that specific sample.
tection limits can not be given. To avoid overloading the chro-matographic system with pesticides, the samples had to be very much diluted. Therefore, in each sample, the detection limit of individual compounds were estimated to be 0.1–5% of the highest found individual concentration depending on the relative response factors of the compounds. For DDT and HCH residues in soil and water the limit of quantification was defined as 10 times the noise.
RESULTSAll samples contained residues of HCHs and DDTs. Particularly high concentrations of the DDTs were found in the surface soil from the old storage site (12 000 to 282 000 mg kg-1 ΣDDT) (Table 1). Soil samples taken at 20–25 and 50–55 cm were lower in concentration compared to surface soil (0–5 cm). The HCHs showed the same pattern with higher concentrations (2300 to 63 000 mg kg-1 ΣHCH) in the surface samples. Surface soil (0–5 cm) sampled 20 m from the old storage site contained levels of DDTs and HCHs similar to surface soil samples from the site itself, but the levels of DDTs in the surface soil sampled 50 m away were lower and all HCHs were below the limits of quanti-fication. Soil samples from the ditch were also high in HCHs and DDTs, up to 7400 and 99 600 mg kg-1, respectively (Table 2). The samples collected downhill from the old storage site were generally more contaminated than the uphill samples. Other pesticides detected in the soil samples were aldrin, azinphos-methyl, γ-chlordane, heptachlor, hexazinone, meta-
mitron, metazachlor and pendimethalin (Ta-bles 1–2). As for HCHs and DDTs, they also showed a patchy distribution at the old stor-age site. One surface soil sample from the old storage area contained 4% of the herbicide pendimethalin. Pendimethalin was also found in some other soil samples. All water samples contained DDTs and HCHs (Table 3). The highest concentrations were found in Well 2 in May 2001. Notably, also the tap water supplied from the water cis-tern contained substantial levels of contami-nants. Other pesticides found in some of the water samples were carbosulfan, chlorprofam, and thiabendazole.
DISCUSSIONThe most important conclusion from this
study is that the problems with abandoned pesticide storage sites may remain for a long time despite attempts to clean up the site. Cleanup in many cases, exemplified by Vikuge, only means removing the visible remains of pesticides and packag-ing. However, if the soil has become contaminated risks to the local environment and population remain. In Vikuge, the soil of the old storage site must be regarded as hazardous waste and a complete degradation of the pesticides still present in the soil will take a long time if nothing is done to remediate the site. As mentioned above, the high levels of DDTs and HCHs made it difficult to find other contaminants. Even so, it is obvi-ous that several other pesticides were also present at high con-centrations (Tables 1 and 2), although the distribution within the contaminated site was even patchier for these than for HCH and DDT residues. The reason is probably that the consign-ment shipped to Vikuge consisted mainly of DDT- and HCH-containing products that thus were spread out over most of the storage area, whereas other pesticides contaminated patches where packages of these happened to be deposited. The pack-aging material present at Vikuge in 1989 (Photos B and C) was from several compounds that were not detected in 2001, most notably some organophosphorous insecticides and telo-drin. The explanation may be i) less persistent pesticides may have degraded to a higher extent than the organochlorines; ii) pesticides more water-soluble than the organochlorines may have infiltrated to groundwater or been washed away; iii) pes-ticides that were present in the consignment in small amounts may have been missed simply because no sample was taken at
Table 2. Concentration (mg kg–1 dry weight) of pesticides in surface (0-5 cm) soil samples from the ditch with the drinking water pipe at Vikuge Farm. Samples were collected in May 2001 uphill and downhill of the old storage site. Figures given as less than (<) indicate presence of the compound, but below the limit of quantification for that specific sample.
Table 3. Concentrations (µg L-1) of pesticide residues in tap water (sampled in August 2000) originating from the water cistern, surface water from a pond, and water from two dug wells (sampled in March and May 2001) at Vikuge Farm. Figures given as less than (<) indicate presence of the compound, but below the limit of quantification for that specific sample.
Tap Pond Well 1 Well 2August (n=1)
March (n=2) May (n=2) March (n=2) May (n=1) March (n=2) May (n=2)
the spot where they were deposited or because the high levels of DDT and HCHs masked their presence; or iv) compounds that are present were not included in the screening (11). Apart from the obvious fourth explanation, explanation three seems the most likely. As an example, telodrin, packages of which were found in 1989, is a highly persistent organochlorine pes-ticide that was manufactured from 1958 to 1965 (12). Existing stocks were used throughout the world for several years after production had ceased, but the agricultural use was restricted due to its high toxicity to mammals and persistence (12). It is noteworthy that telodrin in this case was donated to Tanzania more than 20 years after production ended. We do not yet have sufficient data to make a full environ-mental impact assessment (EIA) of the pesticide contamina-tion at Vikuge such as infiltration to the groundwater, surface runoff during heavy rains, volatilization to air, and wind drift of contaminated dust. However, a few observations are worth mentioning. According to the data at hand, the latter three pro-cesses seem to be the most important, at least for the DDTs and HCHs since the highest concentrations were found in the surface soil. Although most of the DDTs and HCHs in the Vikuge soil were found in the top 0–5 cm of the soil and at a depth of 20–25 cm, it is noteworthy that the organochlorine compounds reached as far as 50–55 cm below the soil surface. The previous inves-tigation by NEMC indicated DDT at a depth of 100 cm (7). At ‘normal’ environmental concentrations this is not expected for compounds as hydrophobic as the DDTs. Possibly, the sorbtive sites in the soil become saturated with pesticides making the downward movement faster than would be expected at lower contaminant concentrations. Another explanation could be that the DDTs and HCHs are transported downwards sorbed to col-loids, a process know as particle-facilitated transport (13). At present, it is impossible to assess the importance of macropores, e.g. cracks, root channels and worm-holes, in the Vikuge soil, for the transport of contaminants to the groundwater (14). Addi-tional investigations of the soil structure are needed to determine the infiltration pathways. DDT and HCH-levels were generally high in the water samples. Samples taken May 7 in Well 2 had a particularly high concentration of DDTs and also the highest levels of HCHs measured in any water samples. A few days prior to the May sampling there had been a heavy, out of season, rain-storm. Well 2 is further away from the most contaminated site than both Well 1 and the pond. However, judging from the topography, the surface runoff from the contaminated site is in the direction of Well 2. The levels of the DDTs were higher than their solubility in water. These high concentrations must be due to high levels of contaminated colloidal particles in the water. Therefore, filtering of the water may be a first step for cleaning the drinking water from contaminants bound to particles. The contaminant concentrations in the water may not seem high considering the very high levels of soil contamination pres-ent for 15 years. However, the data presented here cover only two sampling occasions. An in depth evaluation of the situation requires systematic sampling over at least a year so that seasonal differences can be identified. The overall contamination of the drinking water resources may, therefore, be underestimated in this study, especially if there is substantial surface runoff during the twice yearly rainy seasons. Even the tap water, originating from a presumed clean source, had substantial levels of contaminants. An immediate thought is that the tap water becomes contaminated while passing through the PVC-pipe that runs across the contaminated site. Dust par-ticles entering the cistern storing the water may also be a source
of contamination, and it is also possible that the water source for the cistern water is contaminated. The World Health Organization guideline for DDT residues in drinking water is 2 µg L-1 (15) for protection of human health. This is calculated on the basis of a child (10 kg) drinking 1 L day-1, and DDT exposure through drinking water contributing only 1% of the total exposure. In Vikuge, the exposure through drinking water probably exceeds 1% of the total daily intake, especially if water is not filtered. The WHO guideline for HCHs in drinking water is also 2 µg L-1 (15). The levels of both DDTs and HCHs in the May sample from Well 2 far exceed the guide-lines. Also, the fairly high levels of o,p’-DDT in some water samples are worrying since this DDT-isomer has known estro-genic effects, causing developmental disorders and disturbances in reproductive functions (16). In most samples the relative proportions of the different HCHs clearly represent a technical mixture (17), but some samples differ markedly. Particularly soil samples from the ditch (Table 2) and the water sample collected in May from Well 2 (Table 3), show a high dominance of β-HCH and very little of γ-HCH (lindane), the isomer with insecticidal activity. This indicates that at least part of the consignment consisted of remains of a crude HCH-mixture where the lindane, and thus the insecticidal activity, had been removed for the production of pure lindane. Pure lindane was allowed for use in many Eu-ropean markets long after the technical HCH-mixture had been banned (18). Among the DDT-compounds the relative proportion of p,p’-DDD is too high for the source to be technical grade DDT only (17). Technical grade DDT should, normally, con-tain only a few percent p,p’-DDD, and weathered residues in a well aerated, soil should contain more p,p’-DDE than p,p’-DDD. In these samples the levels of p,p’-DDD are almost equal to, or in some cases even higher than, the levels of p,p’-DDT. However, p,p’-DDD was produced as an insecticide in its own right (15), generally under the acronym TDE (trade name, e.g. Rhothane). Therefore, the reason for the ‘excess’ of p,p’-DDD may be that the consignment contained some formulation with p,p’-DDD as main active ingredient. Since p,p’-DDD has a generally lower efficacy than p,p’-DDT there is a risk of applying too low doses, and thus increasing the risk for resistance development. Therefore, the user must be aware that the active compound is p,p’-DDD instead of p,p’-DDT. Since the documentation is missing it is unclear whether or not the intended users could be informed of this. The produc-tion of p,p’-DDD formulations ceased around 1970 (19), long before this consignment was donated to Tanzania.
CONCLUSIONSAlthough attempts to clean up the site at Vikuge have been made, the site is still highly contaminated. This study shows that to remove the risks to humans and the environment in areas severely contaminated with pesticide, it is necessary to take measures both to dispose of stocked pesticides and to prevent continued environmental contamination from the soil. The presence of superseded compounds, represented by telo-drin and unproportionately high levels of p,p’-DDD, an HCH-mixture from which the active ingredient has been removed, and the presence of empty packaging material in the original shipment, indicate that a main reason for the donation in this case was to get rid of an old stock. Dealing with the problem becomes even more difficult when the documentation of what the shipment contained has been lost and many of the labels are in a language that few read. A case like this stresses the re-
sponsibility of the authorities in donor countries to make sure that donations are relevant and of good quality. At Vikuge, the worst contaminated area is about 30 x 30 m and most of the pesticides were within the top 20 cm soil layer, but some reached a depth of at least 50 cm. Taking this into account, about 180 m3 soil was contaminated with an average DDT concentration of 40 g kg-1. However, it is at present dif-ficult to recommend any immediate action. Our knowledge of the extent of pollution is still incomplete and any measures tak-en might make the situation worse, e.g. by mobilizing bound pesticides and thereby spreading the local contamination even further. It is therefore necessary to initiate actions to solve the acute problem of clean drinking water at Vikuge, and also fur-ther investigate the situation to find the best possible solution both at Vikuge and other sites with similar problems.
References and Notes
1. FAO 1995. Prevention and Disposal of Obsolete and Unwanted Pesticide Stocks in Af-rica and the Near East. Food and Agriculture Organization of the United Nations, First consultation meeting, Food and Agriculture Organization Rome, 43 pp.
2. Guenther, D., Schimpf, W.A. and Vaagt G. 1998. Disposal of obsolete pesticides - joint solutions called for. Pesticide Outlook 9, 5-8.
3. FAO 2001. Baseline Study on the Problem of Obsolete Pesticide Stocks. Food and Agri-culture Organization of the United Nations, Pesticide Disposal Series, 9, Food and Ag-riculture Organization Rome, 42 pp.
4. Van Veen, F. and Breedveld, G.D. 1987. Pesticide Wastes in Zanzibar Islands. Hague, The Netherlands TAUW Infra Consult b.v., 30 pp.
5. Boethling, R.S. and Mackay, D. (eds). 2000. Handbook of Property Estimation Methods for Chemicals: Environmental and Health Sciences. Lewis Publishers, Boca Raton.
6. Wania, F. and Mackay, D. 1996. Tracking the distribution of persistent organic pollut-ants. Environ. Sci. Technol. 30, A390-A396.
7. NEMC 1998. Chemical Waste Management in Tanzania. Report. National Environmen-tal Management Council, Dar es Salaam, Tanzania.
8. Mmochi, A.J. and Mberek, R.S. 1998. Trends in the types, amounts, and toxicity of pesticides used in Tanzania - Efforts to control pesticide pollution in Zanzibar, Tanzania. Ambio 27, 669-676.
9. Åkerblom, M. 1995. Environmental Monitoring of Pesticide Residues. Guidelines for the SADC Region. SADC ELMS Monitoring Techniques Series Vol. 3, Maseru, Leso-tho.
10. FAO 1990. Guidelines for Soil Description. Food and Agriculture Organization Rome, 70 pp.
11. Hexachlorocyclohexane (HCH) has 8 isomers of which we determined the 4 major isomers, α-, β-, γ-, and δ-HCH. The DDTs consist of p,p'-DDT (1,1,1-trichloro-2,2-bis(4-chlorophenyl) ethane), o,p'-DDT (1,1,1-trichloro-2-(2-chlorophenyl)-2-(4chlo-rophenyl) ethane), p,p'-DDD (1,1-dichloro-2,2-bis(4-chlorophenyl) ethane), and p,p'-DDE (1,1-dichloro-2,2-bis(4-chlorophenyl) ethene). ΣHCH and ΣDDT designate the respective sum concentration of these compound groups. The samples were screened for the following pesticides and pesticide degradation products: Acephate, aldrin, at-razine, deethylatrazine, deisopropylatrazine, azinphos-methyl, azoxystrobin, biterta-nol, captan, carbaryl, carbofenothion, carbofuran, carbosulfan, carboxin, α-chlordane, γ-chlordane, chloridazon, chlorobenzilate, chlorothalonil, chlorpropham, chlorpyri-fos, cyanazine, cyfluthrin, λ-cyhalothrin, cypermethrin, p,p'-DDD, p,p'-DDE, o,p'-DDT, p,p'-DDT, deltamethrin, desmedifam, diazinone, dichlorvos, dieldrin, dimetho-ate, diuron, α-endosulfan, β-endosulfan, endosulfan-sulfate, endrin, keto-endrin, ethion, fenmedifam, fensulfothion, fenvalerate, flucythrinate, α-HCH, β-HCH, γ-HCH, δ-HCH, heptachlor, heptachlor-epoxid, hexachlorbenzene, hexazinone, imaza-lil, iprodion, malathion, metalaxyl, metamitron, metazachlor, methabenzthiazuron, methiocarb, methoxychlor, metribuzin, mevinphos, ethyl-parathion, pendimethalin, pentachloroanilin, permethrin, pirimicarb, prochloraz, propiconazol, propoxur, pro-pyzamid, prosulfocarb, quinalphos, quintozene, simazine, sulfotep, telodrin, terbu-tryn, terbutylazine, tetradifon, thiabendazole, tolylfluanid, triadimefon and trifluralin. The samples were also screened for residues of diquat and paraquat with a spectropho-tometric method not described here as no finds were made.
12. WHO 1992. Environmental Health Criteria, No 129: Isobenzan. World Health Organi-zation, Geneva, 62 pp.
13. Villholth, K.G., Jarvis, N.J., Jacobsen, O.H. and de Jonge, H. 2000. Field investigations and modeling of particle-facilitated pesticide transport in macroporous soil. J. Environ. Qual. 29, 1298-1309.
14. Harris, G.L., Nicholls, P.H., Bailey, S.W., Howse, K.R. and Mason, D.J. 1994. Factors influencing the loss of pesticides in drainage from a cracking clay soil. J. Hydrol. 159, 235-253.
15. WHO 1993. Guidelines for Drinking Water Quality. Second edition, volume 1, World Health Organization, Geneva.
16. Colborn, T., Saal, F.S.V. and Soto, A.M. 1993. Developmental effects of endocrine-dis-rupting chemicals in wildlife and humans. Environ. Health Perspect. 101, 378-384.
17. Melnikov, N.N. 1971. Chemistry of Pesticides. Residue Reviews, Vol 36. Springer Ver-lag, Berlin, Heidelberg, New York. 480 pp.
18. Li, Y.-F., McMillan, A. and Scholtz, M.T., 1996. Global HCH usage with 1°x1° longi-tude/latitude resolution. Environ. Sci. Technol. 30, 3525-3533.
19. An exact year when commercial production of DDD-formulations ceased has been dif-ficult to find. The Pesticide Manual published by the British Crop Protection Council (H. Martin, ed.) lists DDD as an actively produced compound in the 1968 edition, while the 1974 edition states that production has ended.
20. This work was supported by the Swedish International Development Cooperation Agen-cy (Sida). Börje Paulsson, Swedish Environmental Protection Agency (formerly at the National Environmental Management Council in Tanzania) has given valuable informa-tion about the situation at Vikuge in 1989. Malin Åkerblom, Henk Bouwman, Sverker Molander, and Rosana Moraes are thanked for valuable discussions.
21. First submitted 18 April 2002. Revised manuscript received 10 Dec. 2003. Accepted for publication 16 Dec. 2003.
Sara Elfvendahl is PhD-student at the Swedish University of Agricultural Sciences. Her research focuses on contamination from obsolete stocks of pesticides in developing countries. Her address: Department of Environmental Assessment, Swed-ish University of Agricultural Sciences, PO Box 7050, SE-750 07 Uppsala, [email protected]
Matobola Mihale holds an MSc degree in chemistry from the University of Dar es Salaam and expects to join the Open University of Tanzania, as an Assis-tant Lecturer. His address: Chemistry Department, University of Dar es Salaam, PO Box 35061, Dar es Salaam, [email protected]
Michael A. Kishimba is senior lecturer at the Univer-sity of Dar es Salaam, Tanzania. His research is on pesticides in the tropics, including residue analy-ses. His address: Chemistry Department, University of Dar es Salaam, PO Box 35061, Dar es Salaam, [email protected]
Henrik Kylin is associate professor at the Swedish University of Agricultural Sciences. He works with the environmental fate of organic contaminants on local, regional, and global scales. His address: De-partment of Environmental Assessment, Swedish University of Agricultural Sciences, PO Box 7050, SE-750 07 Uppsala, [email protected]
