Occurrences of Polar Pesticide Contamination in Niger River Valley and Its Tributary the Mekrou River, Using POCIS Passive Samplers and Survey of Agricultural Practices (Niger Republic)

The increase in food needs due to high population growth in Niger has led to the intensication of urban agriculture and increased use of pesticides. The objective of this study is to assess the polar pesticide contamination of the Niger River and its tributary, the Mekrou River, in Niger, using both grab sampling and POCIS (Polar Organic Chemical Integrative Samplers) and evaluate their ecological risks in the waters. Two water sampling campaigns were carried out during the rainy and dry seasons. The samples were analyzed by liquid chromatography coupled with mass spectrometry (HPLC/MS) and allowed the detection of compounds with concentrations above the WHO guide values and the EU directive: diuron with 3281 ng/L (EU quality guideline: 200 ng/L), atrazine with 1476 ng/L (EU quality guideline: 600 ng/L) and acetochlor with 255 ng/L (EU quality guideline: 100 ng/L). Diuron and Atrazine show a high ecological risk for aquatique environment. The main source of this water contamination is the intensive use of pesticides in urban agriculture near the city of Niamey and in an intensive cotton farming in the Benin. The results of surveys showed the inuence of poor pesticide application practices and seventy percent (70%) of the pesticides used are not approved by the Interstate Committee for Drought Control in the Sahel (CILSS) and some are prohibited in Niger. These contaminations may raise concerns about the health of farmers along the Niger River and the Mekrou River. it for sites F1, F4 F6 for F6.


Introduction
The development of agricultural activities in Africa, occupies an important place for human nutrition and contributes to the national gross domestic product (GDP) of some countries. The growing demand for agricultural products, associated with yields improving involves the systematic use of chemical pesticides to control crop pests In Niger, the annual population growth rate is 3.9%, one of the highest in the world, and agriculture is the country's third largest source of income, after uranium mining and livestock. Agriculture is mainly practiced on small family farms used for food crops (millet, sorghum, cowpea and cassava), rice cultivation, and some cash crops such as corn and peanuts, without recourse to mechanization (MHELD, 2005). It is mainly practiced in the southern part of the country and along the Niger River, and corresponds to about 13% of the national territory (Guengant and Banoin, 2003). Market garden crops such as tomatoes and onions are grown in the depressions created by ancient or recent rivers.
The Mékrou River is a temporary tributary of the right bank of the Niger River that originates in Benin in the plain west of the Atakora Mountains (Le Barbe et al., 1993;Robert Vernet, 1994). This river is heavily involved in cotton production, which began in 1965 and was encouraged by the Compagnie française pour le développement des textiles (CFDT). This production was greatly expanded in the 1980s with the large-scale application of plow-down cultivation (Leo J. De Haan, 1992). Benin is the leading cotton-producing country in Africa, ahead of Mali, Côte d'Ivoire and Burkina Faso, with production of 712,000 tons in 2020 (Benin Ministry of Agriculture), accounting for 80% of the country's export earnings Committee (Zabeirou et al. 2018). The harmful effects of chlorpyrifos ethyl and fenitrothion used against the desert locust on the two species of pimelia (Coleoptera, tenebrioniadae) in the Ta det valley (Agadez region) and occupational exposure to organophosphates and carbamates in rural Niger were highlighted in the work of Mamadou et al., 2005Mamadou et al., , 2008. On the Niger River, the exposure of the diet of the city of Niamey to residues of phytopharmaceutical products during the hot and cold dry season was highlighted and showed that the risk of exposure was higher in children than in adults for all the residues detected and whatever the product (Massalatchi et al. 2018). These exposure risks also concern small farmers with an exposure level above the acceptable exposure level for all active substances and vary from 0.0013 mg / kg bw / day to 0.4125 mg / kg bw / day (Massalatchi et al. 2017). The result of surveys conducted on the risk of respiratory disorders in adults and children related to cultivation in Niger showed that people living in agricultural areas have an increased risk of respiratory symptoms in adults (wheezing, dyspnea, sudden shortness of breath, and cough without fever) and in children (cough without fever) compared to those in pastoral areas (Mamane et al.,2014(Mamane et al., , 2016.
Although the work carried out has highlighted the risks of exposure of certain producers through the use of pesticides, there is currently no work assessing the quality of water to pesticides along the Niger River and on the Mekrou River.
Chemical pesticides are used by farmers to protect their crops from pests and are mostly exported from neighboring countries. The use of these mainly unregistered pesticides for agricultural purposes could be a source of environmental and health problems in the Niger River watershed in Niger. The objective of this study is to study the occurrences of polar pesticides contamination in the Niger River valley and its tributary, the Mekrou River, and to evaluate the inputs of pesticides from the river and its tributaries in the Niger portion of the watershed.

Page 4/27
The passive sampling technique will be used to detect and quantify polar compounds in the water. It has advantages in terms of the preconcentration of compounds in water and increase the possibility of detecting or even quantifying compounds present at very low concentrations in water and improving the diagnosis of contamination (Di Carro et al., 2018). Samplers will be placed along the Niger River and downstream of tributaries to accumulate molecules on a receptor phase by diffusion (mass transfer) of compounds. This technique will be used for the rst time in the assessment in pesticide contamination of the Niger River and its tributaries.

Presentation of the study area and sampling sites
The study area is located in the Niger River watershed, in Niger, on the 550 m, from the border with Mali in the North to the border with Benin and Nigeria in the South, and crosses the region of Tillabéry on 420 km and that of Dosso on 130 km ( Figure 1). Its hydrological regime is highly variable in time and depends on the amount of rainfall in Upper Guinea and Northern Côte d'Ivoire, constituting the Guinean ood in December-January and the local ood between August and September, fed by its tributaries. It is the main watercourse in Niger and its irrigable area is estimated at 142,500 ha (SOGREAH/BRGM, 1981). Several rice schemes are located along the Niger River in both regions, covering an area of about 8,500 ha and are developed by more than 2,000 bene ciaries (Ehrnrooth et al., 2011). Several hydro-agricultural developments (AHA) have been carried out by the State of Niger over an area of 13,000 ha, the management of which was entrusted to the O ce National des Aménagements Hydro-Agricoles (ONAHA), then transferred to farmers' organizations grouped into agricultural cooperatives with the objective of growing twice as much rice per year through the support of ONAHA in the framework of advice and provision of services (Baron et al., 2010).
Niger's climate is Sahelian, characterized by two seasons: a nine-month dry season (October to June) and a three-month rainy season (July to September). Niger's agro-climatic zones are made up of the Sahelo-Sudanian zone, the Sahelian zone, the Sahelo-Saharan zone and the Saharan zone. The Niger River is located in the Sahelian zone with a rainfall of between 350 and 600 mm per year, where agriculture is predominantly rainfed and irrigated. The Mékrou River is located in the most watered part of Niger and crosses Park W, at the common border between Niger, Benin and Burkina. It is covered by the Sahelo-Sudanese zone, which receives 600 to 800 mm of rainfall per year with conditions that are very favorable for rain-fed and irrigated agriculture. The latter receives about 80% of its water from Benin, which is one of the country's main cotton producing areas. The cultivable area of the zone is small so there is no agricultural activity in Park W.

Methodology for the survey of agricultural practices and pesticide use
Information on farming practices and pesticides used by farmers was collected from questionnaires. The representative sample of the surveyed farmer population is composed of 30 producers, mostly men, on the Liboré and Saga hydroagricultural development sites (AHA) for rice production, corresponding to sampling point F4, and the market garden sites located downstream from the city of Niamey along the Niger River. The main agricultural activities practiced along the Niger River are urban agriculture practiced by about 600 market gardeners at the Gamkalle site and the two irrigated perimeters for rice cultivation used by about 1,800 allottees in the surrounding villages. These agricultural activities are practiced for two harvests per year.
The survey questionnaire included general data on the farmer, such as level of education, age, and agricultural training received; data on the characteristics of the farm, such as the area cultivated for market gardening or irrigation, agricultural practices, and the origin of water used; and the type, frequency, and technique of chemical fertilizer and pesticide use. The last part of the questionnaire concerned the integration of producers in professional networks and their knowledge of the environmental issues related to the use of pesticides, as well as their opinion on the prospects for improving their activity.
The objectives of the survey were rst explained to the members of the cooperative and to the producers present on the sites in order to have their involvement; and the survey was carried out with the perfect collaboration of the producers with a response rate of 100%. The survey involved twenty (20) producers from the irrigated sites and ten (10) from the market gardening sites. The preliminary study of the study area according to its size, the types of agricultural activities carried out along the Niger River, in Niger and near its tributaries made it possible to select six (6) sampling sites, ve (5) of which are on the Niger River and one (1) at its con uence with the Mekrou River, in order to investigate the pesticides used and cover the entire watershed ( Figure 1). Site F1 is located at the entrance of the Niger River from Mali, where there is no intense agricultural activity on the Niger side and no input from its tributaries into Niger. Point F2 is located at the con uence of the Niger River with its right bank tributary, the Sirba, where there are agricultural plots for market gardening; and upstream from the pumping stations for the water supply of the city of Niamey. Point F3 is located upstream of the city of Niamey and before the large market gardening sites, and point F4 is positioned downstream of Niamey with several human activities and after the market gardening sites and the irrigated areas of Saga and Liboré. Point F6 is located at the con uence of the Mékrou River with the Niger River, and allows for the evaluation of pesticide contamination from this river; and nally, point F7, located in Gaya, which is a Sahelo-Sudanese area and at the exit of the Niger River from Niger territory, to evaluate pesticide inputs from the Niger side.

Materials
The sampling materials used consisted of a pH meter and a conductivity meter for in situ measurements of physical parameters and 1L and 500 mL glass vials. Filtration was performed using a Nalgene ltration unit (for GL45 ask), a hand pump and GF/F disposable lters. The Oasis HLB cartridges were conditioned at the IMT Mines Alès with 5 mL of acetonitrile under vacuum, followed by 5 mL of methanol (MeOH) and 5 mL of ultrapure water (Ibrahim et al., 2013b) and transported in a cooler to Niger.
For both (2) types of sampling, a membrane pump was used for ltration during the extractions work.

Method
The grab sampling and passive sampling campaigns were carried out during the winter season between August and September and the dry season between April and May. They concerned the six (6) sampling sites selected in order to better evaluate the contamination in the same hydrological context of the Niger River. At each site, in situ measurements were made to collect the pH and conductivity of the environment and the POCIS are deployed in triplicate (3 POCIS) for a period of 15 to 20 days. They are placed in a cage, attached to an empty canister that oats and immersed vertically in river water. Upon removal, the POCIS are rinsed with Milli Q water, wrapped in aluminum foil, then in a plastic bag and stored in a cooler before being transported to the laboratory for extraction.
Water samples are collected manually in 1-liter glass bottles (grab sampling) previously rinsed with the water to be sampled, on the days of deployment and withdrawal of POCIS. They are also kept in a cooler during transport before the extraction phase.

Extraction of water grab samples and POCIS
The extraction phase of the water samples was carried out in Niger. The collected water was divided into two 500 mL asks, and vacuum ltered using a Nalgene ltration unit with a 0,7 µm pore size GF/F lter; in order to remove suspended matter. On each lter, 50 µL of atrazine was added as an extraction yield tracer and the solid phase extraction on Oasis HLB® cartridges. These cartridges were stored in a refrigerator in Niamey before being transported to the laboratory at IMT Mines Alès. Before extraction, the Oasis HLB® cartridges were activated with 5 mL of acetonitrile under vacuum, followed by 5 mL of methanol (MeOH) and 5 mL of ultrapure water (Ibrahim et al. 2013b). The elution phase was performed at IMT Mines Alès after a 1 h vacuum drying. The analytes were then recovered by eluting the cartridges with 8 mL of acetonitrile at a ow rate of 1 mL/min (Mhadhbiet al., 2019). For POCIS, a device was set up in Niamey to transfer the phases into an empty solid-phase tube (SPE) with two polyethylene frits with 20µm porosity at the base and top. A Visiprep solid phase extraction (SPE) collector from Supelco (Bellefonte, USA) was used.
The extracts from the extraction were concentrated under a gentle stream of nitrogen to obtain a nal extract of 1.5 mL, which were spiked with simazine d5 before being analyzed by HPLC-MS/MS (Ibrahim 2013).

Liquid chromatography mass spectrometry analysis of polar pesticides (herbicides)
The analysis of the extracts was performed at IMT Mines Alès by HPLC/MS/MS using an Alliance HPLC system (Waters Series 2695). This system is equipped with a quaternary pump, a degasser and a sample changer. Analytical separation was performed with a Kinetex C18 analytical column (100 mm × 4.6 I.D. × 260; Phenomenex) (Mhadhbi et al., 2019). Milli-Q water and acetonitrile, both including 0.05% formic acid, were used as the mobile phase, at a constant ow rate of 0.4 mL/min. The linear gradient started at 60% and reached 100% in 10 min to allow the passage of acetonitrile, followed by stabilization before returning to initial conditions for 2 min. The system was coupled to a triple quadrupole mass spectrometer (Micromass Quattro micro MT, Waters) equipped with an electrospray ionization (ESI) source, used as a detection device and operated in positive ion mode. Argon was used as collision gas. Ion speci c acquisition of each compound was performed in multiple reaction modes (MRN).
The analytical results were recovered from two transitions, one for the quanti cation of the target substances and one for the con rmation. For POCIS, the sampling rates (Rs), were determined under laboratory conditions for each compound by dividing the slope of the linear regression curve by the average aqueous concentration of selected compounds over a 15-day period (Ibrahim et al. 2012(Ibrahim et al. , 2013. Some of the sample rates were not determined at ITM Mines Alès and were therefore retrieved from the literature (  Flazasulfuron, terbuthylazine hydroxy, and simazine hydroxy were not recovered. A certi ed reference material (WaR™ Pollution Nitrogen Pesticides, Lot No. P246-674) from ERA Waters Company (Golden, USA) was used to determine pesticide recoveries in water samples. This reference material is an aqueous solution composed of 24 pesticides including the 6 pesticides selected in our study (alachlor, atrazine, DEA, DIA, metolachlor and simazine). Pesticide concentrations in the reference material ranged from 3.37 to 16.80 µg/L. The recoveries and coe cients of variation were 108 ± 4% (atrazine), 78 ± 4% (alachlor), 61 ± 3% (DEA), 58 ± 3% (DIA), 110 ± 2% (metolachlor), and 79 ± 2 (simazine) after solid-phase extraction (Oasis HLB) and HPLC/MS/MS analysis (n = 9) of the reference material using the same analytical method.

Statistical analysis of data
For all analyses, triplicate analytical measurements were performed and data were tabulated as mean ± standard deviation. Statistical treatments of the data was performed using STATISTICA analysis software and statistical signi cance was set at p<o.05.

Frequency of detection and quanti cation of herbicides and fungicides in the Niger River
For all the sampling campaigns, twelve pesticides, including ten (10) herbicides composed of acetochlor, alachlor, diuron and its metabolites (DCPMU and DCPU), atrazine and its metabolites (DEA and DIA) atrazin-desethyl-DEA and simazin and two fungicide compounds metalaxyl and tebuconazole were detected, representing 38% of the compounds analyzed and 32% were quanti ed (Figures 2 and 3). For the grab samples, DCPMU and the fungicide tebuconazole were never detected or quanti ed and the fungicide metalaxyl was detected but never quanti ed. DCPU was not detected and quanti ed in POCIS and the herbicide tebuconazole was only detected and quanti ed at low concentrations in POCIS. Alachlor was detected but not quanti ed in any of the samples.
The detection and quanti cation frequencies depend on the types of pesticides and the sampling periods. According to the results of the two sampling campaigns, the POCIS show the highest detection frequencies with 86% for acetochlor, 71% for DIA and 64% for diuron, atrazine and metalaxyl. The highest percentage of quanti cation frequencies (Figure 3) are for POCIS sampled during the rainy season campaign between June and September 2019 and the lowest percentages for both sampling types are between April and May.
During the rainy season, the pesticides with the highest quanti cation frequencies in the grab and passive samplings were diuron with 60%, atrazine with 40% and acetochlor with over 20%. No high concentrations were found for fungicides.
3.2.2 Contamination of the waters of the Niger River and the Mekrou River by herbicides and fungicides

For grab sampling
The highest concentrations, with quanti cation frequencies greater than 20%, were found for the grab sampling. The main pesticides encountered were: diuron (3281±222 ng/L) and atrazine (1476±65 ng/L) with its degradation products, followed by acetochlor (255±9.3 ng/L). The two main pesticides (diuron and atrazine) were measured for the site F6 located at the con uence of the Mekrou River and the Niger River in samples collected during the month of August. The highest concentration of acetochlor was measured for the site F4 during September, at the exit of the city of Niamey.
Pesticide concentrations in the water sampled at F6 decreased over three weeks (August 09-30) and ranged from 3281±222 to 1057±100 ng/L for diuron and from 1476±65 to 814±42 ng/L for atrazine. Atrazine metabolites showed stable concentrations and no variation during the three (3)   The highest concentrations are highlighted in POCIS for site F6 located at the con uence of the Mékrou River and the Niger River and sampled during the month of August. The two main pesticides are diuron with an average concentration of 311 ng/L and 217 ng/L for atrazine. The POCIS for sites F1, F2, F4 and F7 of the two campaigns gave results below the detection limit and therefore not usable.

Summary of results obtained
According to Directive 2008/105/EC, the concentration of acetochlor found for site F4 (255±9.3 ng/L) is two point ve (2.5) times higher than the value of the environmental limit, that of diuron for site F6 (3281 ± 222 ng/L) is fteen (15) times higher than the norms and that of atrazine (1476 ± 65 ng/L) is two (2) times higher than the norm. In all CILSS countries, there is a lack of knowledge of the fate and behavior of pesticides in small streams in agricultural watersheds, particularly due to the great variability of the slope inputs and the complexity of the phenomena involved (Rabiet et al. 2008). The intensive use of pesticides along the Niger River and its tributaries is the anthropogenic factor behind the high contamination of diuron, atrazine and acetochlor.
The high concentrations of atrazine and diuron measured in the waters of the Boumba site (F6) merit special attention, given the large quantities of pesticides used in intensive cotton farming, which is vulnerable to insect and other pest attacks. The Mekrou River has its source 80% in Benin and cotton cultivation is practiced on high altitudes and some of Although there is no data on the intensive use of pesticides in crops in southern Burkina on the Mekrou River, the high concentrations of pesticides found in Benin and Niger in this study show a potential for exposure to environmental and health impacts in this area. The potential impacts of pesticide use can result in decreased soil fertility and release of pollutants; water pollution by nitrate, ammonium NH 4 + and heavy metals (Pb, Zn, Mn) and other toxic compounds as well as acute poisoning, poisoning, decreased fertility and even deaths (Lawani et al. 2017). Table 3 below shows the concentrations observed for sites F4 and F6 during the rainy season.   (3) weeks. This process shows that there is an in uence of rainfall and runoff on the concentration of pesticides in the waters of the Niger River and the Mekrou River (Figure 4 and 5).
The water sample during September, at the end of the rainy season, at point F4 showed very low concentrations of diuron and atrazine. This suggests that the pesticides would have been diluted in the August water stream and a seasonal variation in pesticide concentrations.
The presence of diuron and atrazine with highly variable concentrations at site F6, at the con uence of the Mekrou River and the Niger River, suggests a diffuse contamination of these molecules. The samples with the highest concentrations, during the rst sampling, may be related to point contamination.
The graph Figure 4 shows the distribution of the average annual rainfall in Niger at stations near Niamey, the water level in the Niger River in 2019 and the distribution of the molecules diuron and atrazines analyzed.
The two (2)  In addition to the use of certain unregistered and banned pesticides, certain poor practices by farmers, particularly during the preparation, handling, spreading and storage of pesticides, can also contribute to environmental contamination.
The low concentrations recorded at site F2 show that the Sirba River, which originates in Burkina Faso, does not contribute pesticide pollution to the Niger River. The results of the analysis of the samples from sites F1 and F7 do not show any molecules with concentrations higher than the WHO and EU water quality standards. This shows that the water of the Niger River coming from Mali are not contaminated as well as its outlet to Nigeria. The absence of contamination upstream of the city of Niamey and its presence downstream and after the market gardening sites suggests that there is a link between the intensive use of unregistered pesticides and poor practices in the cultivation system.

Ecological risk assessment
The ecological risk assessment was based on the calculation of ecological risk quotient (ERQ), according to the OECD and European guidelines for diuron and atrazine. The RQ was calculated as a chronic toxicity test on three representative trophic levels which are: sh of species Oncorhynchus mykiss, algae of species Raphidocelis subcapitata and green algae and aquatic invertebrates of species Daphnia magna (Branchet et al. 2018). Daphnia magna species is one of the most commonly used species in ecotoxicological risk assessment. This organism was used in the aquatic ecotoxicology test sheet for the chronic ecotoxicity test (ISO, 10706; OECD, 1998). The Lowest Observed Effect Concentration (LOEC) for each selected pesticide was obtained from the University of Hertfordshire Pesticide Properties Database (https://sitem.herts.ac.uk/aeru/ppdb/en/index.htm). The LOEC corresponds to a 21-day chronic exposure for sh and aquatic invertebrates; and the median effective concentration (EC50) from a 72-h acute exposure was used for algae. The effects studied were growth inhibition for algae, immobilization for aquatic invertebrates and survival for sh (Branchet et al. 2018).
The estimated no effect concentration (PNEC) was calculated according to the formula. PNEC = CMEO / 1000. The value 1000, corresponds to the safety factor applied to take into account the inherent uncertainty of the toxicity data obtained in the laboratory.
The risk quotient value RQ was calculated using the formula RQ = MEC / PNEC with MEC corresponding to the measured environmental concentration of the grab samples (µg/l). The highest concentration in each sample was considered.
The herbicides diuron and atrazine are the two main pesticides identi ed during the contamination survey. The calculated PNEC values for algae, aquatic invertebrates and sh are shown in Table 4 below.
The ecological risk assessment is presented in Table 4, taking into account the concentrations measured at all sampling sites on the Niger River and the Mekrou River. The measured environmental concentrations (MEC) used are the maximum values in each campaign, representing the worst case. show signi cant effects in the microalgae Tetraselmis suecica, notably on a 125% (± 2.3%) increase in doubling time and 25% (± 1.8%) increase in uorescence relative to the presence of reactive oxygen species (ROS, by ow cytometry) and a 25% (± 1.8%) decrease in photosynthetic yield and 38% (± 1.9%) decrease in relative lipid content. The presence of diuron and atrazine, particularly at F4 and F6, constitutes a major ecological risk for the aquatic environment. No ecological risk was identi ed during the dry season and only diuron presents high risks for the three trophic levels.

Conclusion
The present study con rms the pesticide contamination of the waters of the Niger River and the Mekrou River by diuron, atrazine and acetochlor. The high pesticide concentrations in the study area are highlighted during the rainy season mainly at sampling site F6, at the con uence between the Niger River and the Mekrou River and at site F4 located along the Niger River, downstream of the market gardening sites and irrigated perimeters, near the city of Niamey. These sites are the most exposed to pesticide contamination in the Niger River area, but do not represent the integrity of their environment. The origin of this contamination seems to be linked to the increased use of pesticides in urban agriculture on the one hand and in intensive cotton farming in Benin on the other. The pesticide concentrations found could be in uenced by the variation in rainfall, the increase in river ow and the availability of unregistered pesticides in Niger. They are higher during the rst rains at the beginning of the rainy season and dilute with the increase in ow. Sampling with POCIS provided results similar to those of grab sampling and is therefore an appropriate method for environmental monitoring of pesticide contaminants in space and time along the Niger and Mekrou rivers, due to the lack of specialized laboratories for the analysis of organic pollutants, adequate sampling equipment, and the remoteness of some sites. The concentrations of diuron at more than sixteen (16) times the Environmental Quality Standard, atrazine and acetochlor at more than 2.5 times the standard constitutes a worrying situation on the quality of the waters of the Niger River and the Mekrou River. The ecological risk assessment highlighted a high risk for algae (Raphidocelis subcapitata,) aquatic invertebrates (Daphnia) and sh (Oncorhynchus mykiss) for diuron. That of atrazine is also very high for algae and aquatic invertebrates and medium to low for sh.
Given the results of this initial work on pesticide contamination and the high ecological risks for point F6, it is necessary to consider additional sampling along the Mekrou River in the direction of Benin and to set up environmental monitoring of the waters of the Niger River for a good understanding of the variation of contaminants over time and space.

Declarations
Acknowledgements: Sylvie SPINELI, Murielle AVEZAK and Marine BERTRAND for the chemical analyses and Analytical LOD and LOQ calculation Moussa KONATE for the determination of sampling points and advise Availability of data and materials All data generated or analysed during this study are included in this published article. They are openly available.
Declarations: Not applicable The authors con rm contribution to the paper "Occurrences of polar pesticide contamination in Niger River valley and its tributary the Mekrou River, using POCIS passive samplers and survey of agricultural practices (Niger Republic)".
Oumar El Farouk MAMAN ILLATOU contributed to sampling, sampe preparation, survey of agricultural practices and pesticide use, analysis and results interpretation and writing the nal manuscript.
Catherine GONZALES contributed to the program of sampling and sample preparation, analysis and interpretation of results and manuscript preparation.
Marc VINCHES contribute to the preparation of sampling, interpretation of results and manuscript preparation. Figure 1 Location of the study area and sampling sites    Concentration of diuron in relation to water depth and rainfall