Concentration and Distribution of Organochlorine Pesticides in Sediments of the Niger River, Nigeria

at Onitsha, while the lowest concentration (1570±204.5, 1214-1820 µg/kg) was detected in a location at the Nicolas River. Discussion. High values of ∑OCPs (>2000 µg/kg) were detected in all of the locations except in three locations where lower levels were detected. The ∑OCPs were higher during the dry season compared to the rainy season. This may be because the resident time of the sediment transported was higher during the dry season compared to the rainy season, which is characterized by storms, high current, and bottom scour. The chlordane concentration ranged between 24.4 and 134.1 μg/kg dry weight (dw) in locations Nicolas 14 and Lokoja 5; and the Probable Effect Concentration guidelines were exceeded. Dieldrin was detected at very low levels in most of the locations and ranged from 5.67 to 70.3 μg/kg dw in locations Onitsha 9 and 8; and the Probable Effect Concentration guideline was only exceeded in location Onitsha 8; however, the Toxic Effect Concentration guideline was exceeded at all of the locations. Dichlorodiphenyldichloroethane (DDD) was not detected in location Lokoja 6 or 7, and Onitsha 8 and 9, although the concentration in all other locations exceeded the guidelines. Dichlorodiphenyldichloroethylene (DDE) concentrations exceeded the guidelines except in location Nicolas 13. Conclusions. Due to the environmental/human risk and potential danger of the elevated levels of OCPs, there is a need for continuous monitoring of the Niger River. Competing Interests. The authors declare no competing financial interests. (190606) 2019 ©


Introduction
Pollution of aquatic ecosystems is a serious environmental concern, particularly in Africa. The challenge remains a multi-faceted problem due to indiscriminate dumping of waste and illegal use of banned pesticides. 1-3 There is a need for constant assessment and monitoring of hazardous substances, particularly in aquatic environments in developing countries, because rivers are media with transboundary transport of chemical substances.
Several studies have reported the relationships between persistent organic pollutants (POPs) and suspended sediments. Although there is insufficient data on POPs consumption and use in Nigeria, POPs have been widely used in Nigeria on food and cash crops, as well as vector disease control since World War II. 6,7 Data on long range transport, deposition, persistence, bioaccumulation, and toxicity are available for few chemicals, hampering the identification of chemicals that may meet the POPs criteria. 8 Data on POPs are rare in Nigeria, particularly on trends in POP levels in the Niger River ecosystem. Information on the environmental fate of POPs in the Nigerian aquatic ecosystem is needed for appropriate national policy on the management of POPs. The present study aimed to determine the concentration and distribution of organochlorine pesticides in the sediment of the Niger River.

Methods
Sampling locations included the Gurara River (tributary), Lokoja (confluence), Onitsha, Brass, and Nicolas ( Figure  1). Three sites were selected from each location, totaling fifteen representative sites ( Table 1). Collection of samples was done quarterly for 2 years from 2008 to 2009, with the intervals coinciding with the early and late rainy and dry season. A total of 120 samples of sediment (30 samples quarterly) were collected. The distance between each sampling location was approximately 1 km. Depths in the stations ranged between 1.0 to 7.50 m. A motorized boat was used for grabbing at all locations.
Samples of surface sediments were collected at a depth of 15 cm of the surface sediments using a Van Veen sediment grabber. The collection was immediately lifted to the boat. A stainless-steel scoop pre-cleaned with acetone was used to collect about 100 g of the sediment sample into a pre-cleaned 200 ml amber glass bottle. The sample was preserved in a cooler with ice blocks. Samples were transferred and preserved in a deep freezer as quickly as possible to avoid degradation. Samples were stored for the shortest possible time interval between sampling and extraction/ cleanup.

Sample preparation and extraction
The details of sample preparation method are described in the Environmental Protection Agency (EPA) 3570 method and Standwandter and Shutler. 10, 11 Anhydrous sodium sulfate (10.0 g) with 5 g of fresh wet sediment was homogenized in a precleaned mortar and transferred to a pre-cleaned polytetrafluoro ethylene extraction tube. Glass beads (5 to 10) were added, followed by a mixture of acetone (25 mL) and petroleum spirit (1:1); the extraction tube was tightly capped and allowed to stand for a minimum of 20 minutes. Twenty (20) μg/l was then added of the internal standard decafluorobiphenyl in iso-octane. The tube was shaken vigorously until the slurry was freeflowing. The samples were extracted by rotating end-over-end for at least 30 minutes. The solvent layer was filtered through a small glass funnel containing a layer of anhydrous sodium sulfate over a plug of glass wool. The sediment sample was extracted twice more by adding 15 mL of acetone/petroleum spirit mixture. All of the extracts were combined in a round bottom flask of rotary evaporator and the sample volume reduced to about 1.0 mL. The details of the procedure for extraction, extract concentration and extract cleanup are described in Unyimadu et al. 9,12,13 Quality control and data analyses Quality assurance and quality control analyses detailed in Unyimadu et al.
were performed including analysis of procedural blanks. 13 Using the certified reference material from the International Atomic Energy Agency, samples were extracted, cleaned, and analyzed with the same procedure used for the environmental samples. 9,12,13

Gas chromatography analysis
The details of the procedure are described in Unyimadu et al. 9,13 The organochlorine pesticides (OCPs) were analyzed using Hewlett Packard 5890 series II gas chromatography with electron capture detector. The instrument was operated in splitless mode, the oven temperature program started at 90 o C (held for 2 minutes) and was raised 130 o C at 15 o C/min, then to 290 o C at 4 o C/min (holding time 20 minutes). Injector and detector temperatures were 250 and 300 o C, respectively. The flow rate through the column was 3 mL/min -1 . Confirmation of the OCPs was done using a gas chromatograph/mass spectrometer (Shimadzu QP2010) with capillary column type HP1MS (30m x 0.25um x 0.25mm id).

Determination of percent moisture content
The determination of percent moisture content was determined using Equation 1.

MC = (m 1 -m 2 ) / (m 1 -m a ) × 100
Percent moisture content, subtracted The weighing bottles were placed in the drying oven for 1 hour at 105ºC and re-weighed under the same conditions until the weight became constant.

Organic carbon (percent)
Total organic carbon was determined using the back-titration method and calculated using Equation 2. 14 Ten (10) g sediment sample was weighed and transferred to a wide neck 500 mL Erlenmeyer flask and 10 mL of 1 molar solution of potassium dichromate was added. The content of the flask was gently homogenized. Twenty (20) mL of concentrated sulphuric acid was added with a Teflon dispenser. The flask was agitated by rotation for one minute to homogenize the sample (at approximately 120ºC). The flask was placed on an insulating plate and oxidation was allowed to continue for 30 minutes. Two hundred (200) mL of distilled water was added and then 10 mL of phosphoric acid. The flask and its contents were homogenized. Three drops of diphenylamine were added as a titration indicator. The excess dichromate was titrated with 0.5 mol ferrous iron solution (this reagent was freshly prepared each day). The titration was continued until the color changed from purplish blue to a luminous greenish blue. Determination of the end point was facilitated by the addition of 1-2 drops of indicator as soon as the color began to change. 14

Equation 2
Total organic carbon in percent of 105ºC dried sediment = 100 x 3.9 (10-V) / P where P represents the mass in g; V, the volume (mL) of the iron (as Fe 2+ , ferrous ion) solution at a concentration of 0.5 mol/L.

Determination of particle size distribution
The particle size of the sediment samples was determined using the method of Jones et al. 15 The sediment samples were air dried, crushed in a mortar and debris were handpicked. The samples were then weighed. Destruction of organic matter was achieved by treatment of the dried sediment with hydrogen peroxide solution. The destruction of any cement was also carried out by shaking of the sediment with dilute hydrochloric acid.
The soil retained on each sieve was weighed. Each sieve to be used was weighed to 0.1 g. Five hundred (500) g of oven dried sediment was weighed to the nearest 0.1 g. The sediment was sieved through a nest of sieves by handshaking for at least 10 minutes. After shaking, the sieves and the pan were weighed to the nearest 0.1 g with the retained sediment. The weights of sediment retained on each sieve Research were determined. The sum of the weights retained was checked against the original sediment weight. The percentage retained on each sieve was calculated using Equation 3.

Equation 3
(Sediment weight retained on each sieve/original sediment weight) x 100

Results
The target compounds were not detected in the blanks. Spiked sediment samples showed satisfactory precision and recoveries ( Table 2) and Unyimadu, Osibanjo, Babayemi

Table 3 -Particle Size Distribution of Sediment in the Niger River
the detection limit of the equipment was in the range of 0.25 -0.50 ng/l. To monitor the accuracy of the gas chromatography method, the International Atomic Energy Agency standard reference material sediment homogenate (SRM 417) was analyzed with each sample set and the results, based on the standard deviation, were satisfactory. 16

Particle size distribution
The grain size silt percent mean in the Gurara, Lokoja, and Onitsha locations ranged from 6.56-9.89%, 5.85-14.9%, and 6.15-13.0%, respectively; while in the Brass and Nicolas stations, the means ranged from 6.40-31.90% and 50.20-61.90%, respectively, with minima at the Lokoja station and maxima at the Nicolas River location ( Table 3). The total organic carbon content followed the same trend with mean range values in the Gurara, Lokoja, and Onitsha locations of 0.24-0.64%, 0.42-0.80% and 0.41-0.99%, respectively; and in the Brass and Nicolas locations, 0.65-4.20% and 3.15-4.03%, respectively.
Hexachlorocyclohexanes α-hexachlorocyclohexane (HCH), γ-HCH, and δ-HCH were detected in all of the sediment samples analyzed and β-HCH was detected in 96% of the samples ( Table 4). The sequence of concentration in the Niger River stations was β-HCH > δ-HCH > γ-HCH > α-HCH. The distribution of the HCHs varied considerably along the different sampling locations.

Chlordanes
The levels of chlordanes in sediment are shown in Table 5. α-chlordane, γ-chlordane, heptachlor epoxide, and methoxychlor were detected in all the sediment samples analyzed and heptachlor was detected in 84% of the samples. The sequence of concentration in the Niger River stations was methoxychlor > heptachlor > heptachlor epoxide > α-chlordane > γ-chlordane. The distribution of the chlordanes varied markedly along the different locations.

Endosulfans
The concentration of endosulfan in sediment is shown in Table 6. Endosulfan consists of endosulfan I and II isomers. While endosulfan sulfate and the diol are susceptible to photo degradation, endosulfan I and II are resistant. 17 The sequence of occurrence of endosulfan in this study was endosulfan II > endosulfan sulfate > endusulfan I. Endosulfan I was detected in all the samples analyzed, while endosulfan II and endosulfan sulfate were detected in 86% and 81% of the sediment samples, respectively. The concentration of endosulfan II was higher than that of endosulfan I, which can be explained by greater degradation of endodulfan I in the sediment. 18 Endosulfan sulfate, which is a major degradation product of endosulfan, is known to be as toxic as the parent compound.

Dichlorodiphenyltrichloroethanes
DDT, DDE, and DDD were detected in 76%, 100%, and 60% of the sediment samples, respectively ( Table 6). The sequence in the concentration of the metabolites was DDD>DDE>DDT.

Endrin, aldrin and dieldrin
The concentration of endrin in the sediment samples was in the order of endrin ketone > endrin aldehyde > endrin ( Table 7). Endrine and endrine ketone were detected in 100% of the Unyimadu, Osibanjo, Babayemi

Total organochlorine pesticides in sediments
The concentration of the sums of OCPs is shown in

Discussion
Previous studies showed relationships between bound concentrations of PCBs and kepone and tidal cycle. 4,19 Nichols found a positive correlation between suspended-sediment-bound concentrations and the suspendedsediment organic-C content. 4 Abarnou et al. further observed that bound concentrations varied with depths in the water column. Moreover, relationships between total loads of suspended sediment and mean PCB concentrations have been reported. 19 This may suggest differences in the distribution of sediment grain size.

Hexachlorocyclohexanes
The α/γ ratio has been used for identification of the possible sources of HCH. The ratio of α-HCH to γ-HCH > 3 is an indication of an input of technical HCH and long range atmospheric transport and deposition. 20 However, a ratio close Unyimadu, Osibanjo, Babayemi

Table 9 -Seasonal Variation (Mean and Range) of Persistent Organic Pollutants (µg/kg) in the Sediment of Gurara, Lokoja and Onitsha locations of the Niger River
Research to or <1 is characteristic of lindane sources. 21, 22 In the present study, the α/γ-HCH mean ratio of close to 1, with ranges in parenthesis, were observed in the following locations: Gurara 1, 0.77±0.38, (0.14- In other studies in India, the overall ratios ranged between 0.09 to 6.41 with a mean value of 1.96, which reflects the regular usage of technical HCH and lindane (γ-HCH) in these areas.

23-26
Additionally, there is the possibility of transformation of α-and γ-HCH to β-HCH in the environment. 27 Furthermore, the special arrangement of chlorine atoms in the molecular structure of β-HCH makes microbial degradation more difficult than the other isomers, which may lead to the accumulation of β-HCH in sediment. 28 Lower levels of γ-HCH have been reported in marine sediments by Klonova et al., who investigated sediments of James Rose Island and detected levels of 0.20-0.30 µg/kg dw. 29 In the African environment, previous studies also detected lower HCH levels compared to the present study. Sunday detected low levels of α and γ-HCH in Ogunpa, Ona and Oniyere Rivers in Nigeria, the α-HCH concentrations were ND-0.20, ND-0.90, and ND-0.40 µg/kg dw, respectively, while the concentration of γ-HCH in the river sediments were ND-1.20, ND, and ND-2.00 µg/kg dw, respectively. 30 Ojo detected low levels in the Lekki lagoon: α-HCH concentration was ND-116.0 µg/kg dw and γ-HCH was Unyimadu, Osibanjo, Babayemi Organochlorine Pesticides in Sediments of the Niger River, Nigeria

Dichlorodiphenyltrichloroethanes
The possible sources of DDT in the aquatic environment may be identified using the ratios of the parent compound of DDT to DDD or DDE (that is, its metabolites). 38 If the ratio (DDT/ DDD+DDE) is much greater than 1, this may be an indication of fresh use of DDT, while historical application of DDT may be indicated by a smaller ratio (Ma et al.

Endrin, aldrin and dieldrin in sediments
Echols et al. detected low levels of dieldrin and endrin concentrations of 4.80 µg/kg and 3.0 µg/kg, respectively, in sediment sample from the lower Missouri River. 37 Most environmental conditions via biotic or abiotic mechanisms favor the conversion of aldrin to a more persistent product, dieldrin. In most surface water, the presence of aldrin/dieldrin is attributed to particulate surface run-off. The distribution of the aldrins and dieldrins varied markedly along the different study locations. Sunday did not detect aldrin in the river sediments, but low levels of dieldrin were detected in the Ogunpa River (ND-1.80), Ona River (ND-0.50), and Oniyere River (ND-6.0 µg/kg dw). 30

Distribution pattern and seasonal variation
Endrin, chlordane, endosulfan and metabolites were very prominent, compared to DDT, HCH, dieldrin, and isomers which occurred at very low concentrations. The sequence was ∑endrin > ∑chlordane > ∑endosulfan > ∑DDT > ∑HCH > ∑Dieldrin. Seasonal variation ( Table 9) of POPs in the Gurara River and Lokoja location showed marked differences between the rainy season and the dry season. The ∑OCPs were higher during the dry season compared to the rainy season period. This may be because the resident time of the sediment transported was higher during the dry season compared to the rainy season, which is characterized by storms, high current, and bottom scour. The temporal distribution of POPs in the Onitsha location was different from the other locations. In this location, total OCPs were higher during the rainy season compared to the dry season.

Sources of persistent organic pollutants and environmental/human health implications
In Nigeria, OCPs were largely used in agriculture until the 1990s, and OCPS are still being used in some countries against certain insects. 56,57 Some of these pesticides may be persistent in the environment for decades. 3, 58 This persistence may be a result of resistance to physical, chemical, and biological forces of degradation. For instance, the chemistry of β-HCH makes microbial degradation more difficult than the other isomers, which may result in the accumulation of β-HCH in sediment and aquatic organisms, including fish. 28 Marine waters and sediments may be contaminated with OCPs through surface run-off and inflow from rivers. There may also be significant contribution from anthropogenic activities surrounding the river. For example, the River Niger at Onitsha is located in a densely populated area with likely input sources, such as the chemical industry, agriculture, and sewage. 12 Humans may be exposed to OCPs through domestic application of contaminated water and consumption of fish from the river. From the Warri to the Onitsha segments of the Niger River, smoked fish and palm oil/ kernel production are common. 56 Organochlorine pesticides pose problems of human/animal toxicity, slow degradation, fat solubility, and bioaccumulation. 58

Conclusions
The analyses of sediment from the longitudinal section of the Niger River to the Delta revealed a wide spectrum of persistent organic pollutants. The sources of most of the pollutants could be attributed to agricultural and industrial activities, waste water from slaughter houses, and storm water run-off. These compounds consist of agricultural, industrial sewage and waste derived particularly from slaughter houses along the banks of the River. Their ubiquitous occurrence has been confirmed by other authors in other Nigerian Rivers. 59 In the sampled sediment, endrin, chlordane, endosulfan, and DDTs were dominant compared to HCHs and dieldrin. The consensus-based sediment quality guidelines 60 were exceeded in all locations (Table 10) for dieldrin, chlordane, DDD, DDE, DDT, heptachlor epoxide, and lindane, sometimes 5-to 10-fold. 61, 62 The ∑OCPs were higher during the dry season compared to the rainy season period. This may be because the resident time of the sediment transported is higher during the dry season compared to the rainy season, which is characterized by storms, high current, and bottom scour.