Chemosphere 72 (2008) 21–24

Contents lists available at ScienceDirect

Chemosphere
journal homepage: www.elsevier.com/locate/chemosphere

Persistent organochlorine pesticide residues in fish, sediments and water from Lake
Bosomtwi, Ghana
Godfred Darko *, Osei Akoto 1, Caleb Oppong 2
Department of Chemistry, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana

a r t i c l e i n f o a b s t r a c t

Article history: Some organochlorine pesticide residues in tilapia fish (Tilapia zilli), sediment and water samples from
Received 21 November 2007 Lake Bosomtwi (the largest natural lake in Ghana) were determined to find out the extent of pesticide
Received in revised form 22 February 2008 contamination and accumulation in the lake. The extracted residues were analyzed on a micro-capillary
Accepted 25 February 2008
gas chromatograph equipped with electron capture detector. DDE (p,p0 -1,1-dichloro-2,2-bis-(4-chloro-
Available online 7 April 2008
phenyl)ethylene) was the predominant residue in all the samples analyzed; detected in 82% of water
samples, 98% of sediment samples and 58% of fish samples at concentrations of 0.061 ± 0.03 ng g1,
Keywords:
8.342 ± 2.96 ng g1 and 5.232 ± 1.30 ng g1, respectively. DDT (p,p0 -1,1,1-trichloro-2,2-bis-(4-chloro-
Fish
Gas chromatography
phenyl)ethane) was detected in 78% at a mean concentration 0.012 ± 0.62 ng g1 of water samples
Organochlorine analyzed. The mean concentrations of DDT in sediments and fish were 4.41 ± 1.54 ng g1 and
Pesticides 3.645 ± 1.81 ng g1, respectively. The detection of lower levels of DDT than its metabolite, DDE, in the
Sediments samples implies that the presence of these contaminants in the lake is as result of past usage of the
Water pesticides.
Ó 2008 Elsevier Ltd. All rights reserved.

1. Introduction dences of their continuous usage in many developing nations, like

Organochlorine pesticides (OCs) are among the agrochemicals
that have been used extensively for long periods in Ghana
(Clarke et al., 1997). Lindane (c-1,2,3,4,5,6-hexachlorocyclohexane),
aldrin (1,2,3,4,10,10-hexachloro-1,4,4a,5,8,8a-hexahydro-1,4,5,8-
dimethanonaphthalene), endosulfan (6,7,8,9,10,10-hexachloro-1,5,
5a,6,9,9a-hexahydro-6,9-methano-2,4,3-benzo(e) dioxathiepin-3-
oxide), dieldrin (1,2,3,4,10,10-hexachloro-6,7-epoxy-1,4,4a,5,6,7,8,
8a-octahydro-1,4,5,8-dimethanonaphthalene), and DDT (p,p0 -1,1,
1-trichloro-2,2-bis-(4-chlorophenyl)ethane) have been used widely
in agriculture, as well as, in mosquito, termite and tsetsefly control
programmes. Many of these OC and their metabolites have been
implicated in a wide range of adverse human and environmental
effects including reproduction and birth defects (Edwards, 1987),
immune system dysfunction, endocrine disruptions, and cancer
(WWF, 1999).
The use of organochlorine pesticides for agricultural purposes
has been banned for about 25 years now in the developed world
and some developing nations like Ghana. However, there are evi-
* Corresponding author. Tel.: +233 24 3239354; fax: +233 51 40808/60305.
E-mail addresses: godfreddarko@yahoo.com (G. Darko), wofakmann@yahoo.com
(O. Akoto), caleboppong@hotmail.com (C. Oppong).
1
Tel.: +233 208216685; fax: +233 51 60305.
2
Tel.: +233 244123816; fax: +233 51 36165.
Ghana, due to inadequate regulation and management on the pro-
duction, trade and use of these chemicals (Darko and Acquaah,
2007). Residues and metabolites of many OCs are very stable, with
long half lives in the environment (UNEP, 2002). Studies have
shown that DDT is still in its highest concentration in biota of some
developing countries that have banned the use of OCs some 20
years ago (Doong et al., 2000; Zhang et al., 2002). Even in the devel-
oped nations, that effectively enforced the ban on the use of orga-
nochlorine pesticides, DDT and OC residues are still detected in
their waters (Hung and Thiemann, 2002). Many other recent works
have indicated the presence of OC residues in surface waters, sed-
iments, biota and vegetations in Africa (Dem et al., 2007; Ize-Iyamu
et al., 2007; Ntow, 2005) and elsewhere (Haozheng et al., 2007;
Imo et al., 2007). The persistent nature of organochlorine residues
in the environment may pose the problem of chronic toxicity to
animals and humans via air, water and foods intake.
Lake Bosomtwi is the largest natural lake in west Africa. It is lo-
cated in the west central part (06° 320 N and 01° 250 W) of Ghana’s
Ashanti Region, 35 km south-east of Kumasi – Ghana’s second larg-
est city. The lake is one of the six major meteoric lakes in the world.
It has a rim diameter of 10.5 km with four streams flowing into it;
but the lake lacks an outlet (Boamah and Koeberl, 2007). Lake
Bosomtwi is a source recreational water as well as a source of
cheap, affordable protein in the form of tilapia fish (Tilapia zilli).
The inhabitants around the lake are predominately vegetable crops

0045-6535/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.chemosphere.2008.02.052
22 G. Darko et al. / Chemosphere 72 (2008) 21–24
farmers who use pesticides in their farming activities. In spite of
the high dependence of the tilapia fish in the lake as a source pro-
tein in the region, no published work is available on the levels OC
contaminates in the biota of the Lake Bosomtwi. Also, judging by
the large usage of pesticides coupled with low perception of the ef-
fects of these chemicals among Ghanaians (Amoah et al., 2006),
there is the need to monitor the levels of these hazardous chemical
in the lake’s biota.
The determination of OC residues in fish, sediments and water
may give indication of the extent of aquatic contamination and
accumulation characteristics of these compounds in the tropical
aquatic biota that will help in understanding the behaviour and
fate of these persistent chemicals (Kannan et al., 1995). This work,
therefore, seeks to provide baseline information on levels of pesti-
cide residues in tilapia fish (Tilapia zilli), sediments and surface
waters of Lake Bosomtwi that will assist in a scientific assessment
of the impact of pesticides on public health, agriculture and the
environment in Ghana.
2. Methodology
All solvents were purchased from BDH and were of pesticide
residue grade. Organochlorine pesticides standards (Lindane, al-
drin, dieldrin, p,p-DDE, p,p-DDT, and endosulfan) were purchased
from Ehrenstorfer, GmbH, Germany, in sealed vials. Solid Phase
Extraction (SPE) bond elut C-18, 3 cc/500 mg were purchased from
Varian Inc., USA. Glassware was washed with detergent, rinsed
with purified water and acetone and were heated to 180 °C for 2 h.
A total of 50 samples each of fish (Tilapia zilli), sediments and
water were collected randomly around the lake. Sampling was
done in two batches of 10 samples each between the period of
May–July 2004, for the first batch, and December 2005–March
2006, for the second batch. Sampling was done weekly.
Water samples (1.5 l) were collected into 2 l cleaned glass bot-
tles from midstream positions of the lake at approximately 0.3 m
depth below the water surface. Water samples were pre-filtered
through 0.45 lm fiber glass filters (Whatman) to remove debris
and suspended materials. Samples were preserved by adding
5 ml of concentrated H2SO4 to prevent biological activity (Ntow,
2005). Waterbed sediments (1 l) were grabbed from the bottom
of the lake, where fine-textured substrate had accumulated, into
glass beakers. Fish samples obtained from fishermen at the various
sites around the lake were washed in enough distilled water. All
samples collected (water, fish & sediments) were immediately
stored in an ice-chest at 4 °C and transported to the laboratory
for analysis.
In the laboratory, the muscle tissues of the fish samples were
ground in a waring blender to obtain a homogenous composite
while the sediment were air-dried. Pre-filtered water samples
(1 l) were processed using a Solid Phase Extraction (SPE) technique
(Totolin, 2003). The SPE cartridges (C-18) were first conditioned
with 5 ml of methanol followed by 5 ml of acetone and 5 ml of
Milli Q water. The water samples (1 l) were passed through pre-
Table 1
Mean Recovery (%), relative standard deviation (RSD) (%), limits of detection (LD) (n = 5)
Water Sediment
1
Recovery RSD (%) LD (ng g ) Recovery
Lindane 85 7 0.50 80
Endosulfan 78 8 0.25 76
Aldrin 91 6 0.30 85
Dieldrin 89 5 0.30 80
p,p-DDE 92 4 0.10 95
p,p-DDT 97 5 0.10 90
conditioned SPE cartridges at a flow rate of 5 ml min1. The sta-
tionary phase washed with 5 ml of acetone and 5 ml of Milli Q
water while wet and then centrifuged at 1000–1500 rpm for
5 min. The analytes trapped in the cartridges were eluted 3 times
with 5 ml portions of hexane. The extracts were concentrated to
about 2 ml using a Buchi rotary evaporator operating at 6 mbar
and temperature of 30 °C.
OC residues in sediments and fish samples were extracted using
Soxhlet Extraction (Therdteppitak and Yammeng, 2003). A 10 g
sample was placed into a beaker containing 50 g anhydrous so-
dium sulfate and mixed thoroughly. The sample mixture was
transferred to an extraction thimble, and placed in a Soxhlet
extractor. The mixture was extracted with 150 ml of acetone: n-
hexane (20:80 v/v) at 50 °C for 4 h. The extracts were filtered, con-
centrated to 1 ml using vacuum rotary evaporator. Each of the raw
extracts was then dissolved in 10 ml hexane and passed through
pre-conditioned octadecyl C-18 columns at a rate of 2 ml min1
to clean up. The column was washed with 1 ml, 30% methanol fol-
lowed by 1 ml ultrapure water and was allowed to dry. The sample
(analyte) which was trapped in the column was eluted 5 times
with 0.5 ml aliquots of hexane to recover the pesticide residues.
Hexane in the sample was then allowed to evaporate off leaving
the residue alone in the vial. Dried sample was dissolved in 1 ml
portion of hexane, mixed thoroughly with a whirl mixer and then
transferred to autosampler vials ready for gas chromatography.
The OCs residues were analyzed using Shimadzu Gas Chromato-
graph, GC-9A, that is equipped with a 63Ni Electron Capture Detec-
tor and SPB-608 (30 m  0.32 mm i.d.  0.25 m film) fused silica
capillary column. The gas chromatographic analysis was per-
formed under the following conditions: the detector temperature
was 320 °C. The injector temperature was 220 °C and the column
temperature was 160 °C (isothermal). Carrier gas was helium at a
flow rate of 30 cm s1. One micro litre (1 ll) of sample was injected
into the GC.
Recovery analyses were carried out on samples fortified at
1 ng g1. After extraction and solvent evaporation, the samples
were analyzed according to the proposed method. The recovery
values were calculated from calibration curves constructed from
the concentration and peak area of the chromatograms obtained
with standards of the organochlorine pesticides (Table 1). Detec-
tion limits of the method were found by determining the lowest
concentrations of the residues in each of the matrices that could
be reproducibly measured at the operating conditions of the GC
(Table 1). Blank analyses were also performed in order to check
interference from the sample. All analyses were carried out in trip-
licates and the mean concentrations were calculated based on the
total number of each sample.
3. Results and discussions
Results from the study (Table 2) have shown that OC residues
are present at low concentrations in sediments, fish and water
samples from Lake Bosomtwi.
Fish
1
RSD (%) LD (ng g ) Recovery RSD (%) LD (ng l1)
8 0.55 80 13 0.70
4 0.52 78 5 0.60
10 0.25 81 7 3.00
12 0.18 79 5 2.00
5 0.24 92 11 2.00
7 0.26 95 8 3.00
 G. Darko et al. / Chemosphere 72 (2008) 21–24 23

Table 2
Levels of pesticide residues in samples

Water (n = 50) Sediment (n = 50) Fish (n = 50)

1 1 1 1
Mean (ng g ) SD Range (ng g ) Percent Mean (ng g ) SD Range (ng g ) Percent Mean (ng g1) SD Range (ng g1) Percent
Lindane 0.071 1.85 0.751.61 64 6.755 1.15 2.0313.94 68 0.126 0.11 0.701.36 38
Endosulfan 0.064 0.24 0.330.42 56 9.683 1.76 3.7514.40 42 0.713 0.94 0.842.32 26
Aldrin <LD – – 0 0.065 0.04 0.300.46 20 0.018 0.35 0.300.49 16
Dieldrin <LD – – 0 0.072 0.02 0.200.46 32 0.035 0.42 0.300.56 38
p,p-DDE 0.061 0.03 0.100.90 82 8.342 2.96 11.5226.40 98 5.232 1.30 4.107.25 58
p,p-DDT 0.012 0.62 0.130.86 78 4.41 1.54 3.545.52 70 3.645 1.81 3.404.65 66

<LD = below detection.

SD = standard error of mean.

Lindane was detected in 64% of water samples, 68% of sediment
samples and 38% of fish samples analyzed with the highest concen-
tration of 13.94 ng g1 occurring in sediments. This result suggests
lindane is more prevalent and persistent in the sediments than in
water and then in fish. The average concentration of lindane in sed-
iments was about 50-folds the average concentration in fish and
90-folds the average concentration in water. It is therefore ex-
pected that lindane levels in fish will rise with time as they utilize
the water and the sediments in the lake.
Endosulfan (marketed as Thiodan), a broad spectrum contact
insecticide and acaricide, is another pesticide used by many Gha-
naian subsistence farmers. The highest concentration of endosul-
fan (14.4 ng g1) was detected in the sediment samples, however,
the highest occurrence of detectable concentrations was found in
water (56%).
Endosulfan level in fish, on the average, is 10 times that found in
water samples. This suggests that the residues in the water are
accumulated and concentrated in the fish. Mean level of endosul-
fan in the sediments is about 14 times the levels measured in fish.
Bioconcentration of endosulfan in fish could therefore arise.
Neither aldrin nor dieldrin was detected in the water samples.
Aldrin and dieldrin residues measured in fish and sediment sam-
ples are therefore as result of the past usage of the pesticides. Orga-
nochlorine pesticides are less soluble in water. They, therefore,
accumulate in fishes and sediments when they are discharged into
water bodies. Frequency of detection (incidence) and concentra-
tion of aldrin measured in sediments were higher than those re-
corded for fish samples. Aldrin was detected in 16% of fish
samples at concentrations ranging from >0.3 to 0.49 ng g1. Aldrin
was detected in 20% of the sample and their concentrations ranged
from 0.25 to 0.33 ng g1. The incidence ratio of aldrin in sediments
was 20% and its concentration ranged between 0.25 and
0.33 ng g1. Dieldrin was detected at an average concentration of
0.035 ng g1 and 0.072 ng g1 in sediments and fish, respectively.
Levels of aldrin in fish and in sediments were lower than those re-
corded for its metabolite, dieldrin. This gives an indication that the
dieldrin residues present are not a result of decomposition of aldrin
but direct application and bioaccumulation of dieldrin as an
insecticide.
DDT was detected in 78% of fish samples, 70% of sediments and
66% of water samples at mean concentrations of 0.012 ng g1,
4.41 ng g1 and 3.645 ng g1. The ratios of incidence as well as
the concentrations of DDE, a metabolite of DDT, in all the three
sample types were higher than those recorded for DDT. This ob-
served trend could be attributed to the decomposition and bioac-
cumulation of the DDT used in the past. Other isomers of DDT
were not analyzed for, due to unavailability of standards.
Ntow (2005) reported average lindane concentrations of
0.008 lg l1 (8 ng g1) in water, and 2.30 lg kg1 (2.3 ng g1) in
sediments sampled from the Volta Lake, the largest lake in Ghana.
These results agree with the 7.10 ng g1 and 6.56 ng g1 lindane
recorded for water and sediments from Lake Bosomtwi in this
work. However, DDE concentration of 52.30 lg kg1 recorded for
sediments from the Volta Lake is about seven folds that recorded
for Lake Bosomtwi. The marked difference in DDE levels from these
two Ghanaian lakes may be due to the fact that all major rivers
along which most commercial farming activities in the country
take place lead to the Volta Lake. Furthermore, Lake Volta, unlike
Bosomtwi, experiences bad fishing practices like use of chemicals
and explosives. Agrochemicals are not extensively used at the ba-
sin of Lake Bosomtwi, since agricultural practices are non-chemical
because farmers are unable to afford extensive use of agrochemi-
cals (Boamah and Koeberl, 2007). The results obtained in this study
generally agree with those obtained from Lake Tanganyika (Mani-
rakiza et al., 2002) and Lake Victoria (Kasozi et al., 2006) – two
other major lakes in Africa.
The levels of most of the residues in fish were higher than those
found in water. Organochlorine pesticide residues in the lake are
likely to originate from nonpoint sources via runoff, atmospheric
deposition, and leaching due to agricultural applications and vector
control practices. The lake sediments act as a sink for the persistent
contaminants, whose resuspension during the lake’s mixing may
increase pesticide bioavailability and accumulation in the fish. Pes-
ticide pollution to the lake is therefore, likely to pose a danger to
both aquatic organisms and humans.
Acknowledgements
The authors would like to express their gratitude to Dr. Maria
Teresa Baeza Romero (School of Earth and Environment, University
of Leeds), Prof. Kurunthachalam Kannan (Environmental Health
Sciences, Wadsworth Center, State Department of Health, USA),
Dr. William Oduro (Inorganic Chemistry Laboratory, Oxford Uni-
versity) and Dr. Leonid Sasraku-Neequaye (School of Chemical Sci-
ences and Pharmacy, University of East Anglia, Norwich) for their
support in various forms towards execution of this project.
References
Amoah, P., Drechsel, P., Abaidoo, R.C., Ntow, W.J., 2006. Pesticide and pathogen
contamination of vegetables in Ghana’s urban markets. Arch. Environ. Con. Tox.
50, 1–6.
Boamah, D., Koeberl, C., 2007. The Lake Bosumtwi impact structure in Ghana: a brief
environmental assessment and discussion of ecotourism potential. Metorit.
Planet Sci. 42, 561–567.
Clarke, E.E.K., Levy, L.S., Spurgeon, A., Calvert, I.A., 1997. The problems associated
pesticide use by irrigation workers in Ghana. Occup. Med. 47, 301–308.
Darko, G., Acquaah, S.O., 2007. Levels of organochlorine pesticides residues in meat.
Int. J. Environ. Sci. Tech. 4, 521–524.
Dem, S.B., Cobb, J.M., Mullins, D.E., 2007. Pesticide residues in soil and water from
four cotton growing areas of Mali, West Africa. J. Food Environ. Sci. 1, 1–12.
Doong, R.A., Sun, Y.C., Liao, P.L., Peng, C.K., Wu, S.C., 2000. The distribution and fate
of organochlorine pesticide residues in sediments from the selected rivers in
Taiwan. Chemo. 48, 237–246.
Edwards, C.A., 1987. The environmental impact of pesticides. Parasitis. 86, 309–329.
Haozheng, W., Mengchang, H., Chunye, L., Xiangchun, Q., Wei, G., Zhifeng, Y., 2007.
Monitoring and assessment of persistent organochlorine residues in sediments
from Daliaohe river watershed, Northeast of China. J. Environ. Monitor. 133,
231–242.
24 G. Darko et al. / Chemosphere 72 (2008) 21–24
Hung, D.Q., Thiemann, Q., 2002. Contamination by selected chlorinated pesticides in
surface waters in Hanoi, Vietnam. Chemo 47, 357–367.
Imo, T.S., Sheikh, M.A., Hirosawa, E., Oomori, T., Tamaki, F., 2007. Contamination by
organochlorine pesticides from rivers. Int. J. Environ. Sci. Tech. 4, 1–9.
Ize-Iyamu, O.K., Asia, I.O., Egwakhide, P.A., 2007. Concentrations of residues from
organochlorine pesticide in water and fish from some rivers in Edo State
Nigeria. Int. J. Phys. Sci. 2, 237–241.
Kannan, K., Tanabe, S., Tatsukawa, R., 1995. Geographic distribution and
accumulation features of organochlorine residues in fish in tropical Asia and
Oceania. Environ. Sci. Tech. 29, 2673–2683.
Kasozi, G.N., Kiremire, B.T., Bugenyi, F.W.B., Kirsch, N.H., Nkedi-Kizza, P., 2006.
Organochlorine residues in fish and water samples from Lake Victoria, Uganda.
J. Environ. Qual. 35, 584–589.
Manirakiza, P., Covaci, A., Nizigiymana, L., Ntakimazi, G., Schepens, P., 2002.
Environ. Pollut. 117, 447–455.
Ntow, J.W., 2005. Pesticide residues in Volta Lake, Ghana. Lake Reserv. Manage. 10,
243–248.
Therdteppitak, A., Yammeng, K., 2003. Determination of organochlorine pesticides
in commercial fish by gas chromatography with electron capture detector and
confirmation by gas chromatography – mass spectrometry. Science Asia 29,
127–134.
Totolin, M., 2003. The extraction and clean up of organochlorine pesticides and
PCB’s from water using solid phase extraction and gas chromatography
analysis. Environ. Eng. Manage. J. 2, 25–28.
UNEP, 2002. Sub-Saharan Africa, Regionally Based Assessment of Persistent Toxic
Substances. United Nations Environment Programme, Chemicals (UNEP
Chemicals), Geneva, Switzerland, p. 118. <http://www.chem.unep.ch/Pts/
regreports/ssafrica.pdf>.
WWF (World Wildlife Fund), 1999. Hazards and exposures associated with DDT and
synthetic pyrethroids used for vector control. World Wildlife Fund, Washington,
DC.
Zhang, Z., Hong, H., Zhou, J., Wang, X.J., 2002. Transport and fate of organochlorine
pesticides in the River Wucuan. Southeast China. Environ. Monit. 4, 435–441.