Physico-chemical and bacteriological quality of groundwater in a rural area of Western Niger: a case study of Bonkoukou

The precariousness of the rural population in Africa is often symbolized by the lack of potable and safe drinking water. This study investigates the physico-chemical and bacteriological characteristics of 32 water samples with respect to WHO standards. The water samples were collected from wells, boreholes and small drinking water supply systems (DWS) in and around the township of Bonkoukou (Niger). The Water Quality Index (WQI) tool was used to assess the overall water quality with different physico-chemical parameters. Where the pH of the samples was acceptable, the samples showed higher levels of mineralization and deoxygenation. Overall, the samples were slightly hard, chlorinated and sulfated but much alkaline and contained nitrate and nitrite ions 2–16 times higher than the WHO standards. The use of WQI shows that samples in the DWS are safe for drinking. Samples coming from wells are the most polluted (58.50%) compared to those taken from boreholes (53.00%), while the percentage of samples from boreholes, unfit for drinking, is higher (41.00%) than that of the samples taken from wells (25.00%). Moreover, water in this area was characterized by the presence of total germs indicating bacteriological pollution. Hence, for the supply of safe drinking water to the larger number of people in such a rural area, the capacity of actual DWS must be improved and widespread. This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY 4.0), which permits copying, adaptation and redistribution, provided the original work is properly cited (http://creativecommons.org/licenses/by/4.0/). doi: 10.2166/wh.2020.082 s://iwaponline.com/jwh/article-pdf/18/1/77/651413/jwh0180077.pdf Hassane Adamou (corresponding author) Rabani Adamou Département de Chimie, Université Abdou Moumouni, P.O. Box 10662, Niamey, Niger E-mail: ada_hassa@yahoo.fr Boubacar Ibrahim Département de Géologie, Université Abdou Moumouni, P.O. Box 10662, Niamey, Niger Seyni Salack Safietou Sanfo WASCAL, Competence Center, P.O. Box 9507, Ouagadougou, Burkina Faso Stefan Liersch Potsdam Institute for Climate Impact Research, RD2 Climate Resilience, P.O. Box 60 12 03, Potsdam, Germany


INTRODUCTION
The main use of drinking water is to compensate for body water loss and to ensure good general physiology and health (Mann & Truswell ). According to UNICEF (), everyone has the right to have access to safe drinking water. Potable and safe drinking water brings clear health benefits. Water that is unsuitable for nutrition, on the other hand, represents serious health risks. According to WHO (), about 30.00% of the world population still has no access to safe drinking water. On the African continent, 320 million people remain excluded from drinking water services (Bazié ), where in West Africa, drinking water coverage is below 60.00% (CMAE ). In a country like Niger, the rate of access to drinking water in rural areas is theoretically estimated at about 46.00% (MHA ). In rural West Africa, groundwater is very often the main source of drinking water. The latter is supplied mostly through wells and increasingly from boreholes (Chippaux et al. ; Raphael et al. ), sparsely distributed across rural areas townships and villages. In addition to the low spatial coverage of the groundwater pumping infrastructure, rural communities face also the challenges of population growth and the degradation of the existing hydraulic structures due to poor maintenance. The reality, in many rural townships and villages in the rural Niger, is that national companies in charge of drinking water supply do not exist.
To compensate this gap, water is supplied in townships and villages, like our study site of Bonkoukou (Western Niger, 140 km northeast of Niamey, the capital city), through wells, boreholes and small drinking water supply systems (DWS). These DWS carry water from a water tower to households through distribution pipes and taps.
However, such small DWS cover only a small portion of the rural areas in Niger.
Water availability alone does not safeguard nutrition nor does it guarantee an access to healthy water. In different parts of the world, particularly rural areas, several research studies have been conducted to assess the quality of drinking water (Mulamattathil et

Study area and sampling points
In Niger Republic, the Tillabéry region has the densest hydrographic network of the country, where the River Niger is the only river with the permanent flow over 400 km across the western parts of the territory. The surface hydrology is characterized by a strong endorheic system. ship is located at 140 km northeast of Niamey (Niamey is the capital city of Niger Republic) with ∼20,000 inhabitants spread across more than 40 villages. The economy of these communities is predominantly based on rainfed farming of millet, cowpea, off-season small-scale irrigation of potatoes, livestock breeding (e.g. sheep, cattle, etc.) and inter-village trading through weekly markets. The ground water in Bonkoukou is shallow, supporting thereby small-scale irrigation and enabling the many local people to practise market gardening.
In addition to the township of Bonkoukou, the three surrounding villages (Djami, Beba Tondi and Mokirey) were included in the water sampling process. Some of the water samples were taken from sources constructed to supply drinking water and other samples came from sources designed to supply irrigation water. A total of 32 samples were collected, of which 26 (81.25%) were taken in the town of Bonkoukou and six (18.75%) in the surrounding villages. The water was sampled from a water tower, two taps, 12 wells and 17 boreholes ( Figure 1). Water from the boreholes and the water tower (WT) is exclusively used for drinking and other household uses, the wells are mostly used for other activities (laundry, showers, etc.), animal watering and small irrigation.

In situ parameters measurement and laboratory analyses
The water samples were collected in July 2018 following the approach described by Rodier et al. (). Sterilized 500 mL polypropylene vials were used to conserve the water samples before the chemical and bacteriological analysis. These samples were stored under a temperature of 4 C to avoid changes in parameters and were quickly analyzed in the laboratory. During the sampling, a 3430 SET F, WTW multi-function multimeter was used to directly measure the physical parameters (temperature (T), hydrogen potential (pH), electrical conductivity (EC), total dissolved solids (TDS) and dissolved oxygen (DO)).

Chemical analysis
The following ions CO 2À 3 , HCO À 3 , Ca 2þ , Mg 2þ , Cl À , NO À 3 , NO À 2 , Fe 2þ and SO 2À 4 were tested and used to determine the chemical quality of the water samples.  These analytical methods are sensitive, simple, cheap and easily applicable. They also give very good results.
UV-Visible spectrophotometric methods are characterized by excellent detection ranges from 0.25 to 20,000 μg L À1 with a recovery rate almost equal to 100%.

Bacteriological analysis
Bacteriological quality is one of the important parameters of water potability. It is measured by the presence of a pollution indicator of organisms, in particular, total germs and fecal coliforms (Escherichia coli). Total germs represent the density of the bacterial population in drinking water. The results are analyzed by considering the bacteriological standards for the detection and enumeration of fecal bacteria in water of the WHO and the International Organization for Standardization (ISO). With regard to these standards, in any water (directly intended for drinking, treated and entering the distribution system), any E. coli or thermotolerant coliform bacteria must not be detectable in a volume of 100 mL (WHO ).

WQI analysis
In this study, water quality (physico-chemical and bacteriological) and its degree of pollution were assessed in relation to the standards defined and recommended by the World Health Organization (WHO ). In addition, the WQI was used in the assessment as the practical tool for synthesizing physico-chemical water quality parameters (CCME ). The WQI is commonly used for the detection and assessment of water pollution and can be defined as the reflection of the composite influence of different quality parameters on overall water quality. The use of WQI makes more comprehensible the meaning of the analyzed parameters. The calculation of the WQI was based on the method described by Batabyal & Chakraborty (). A correlation (r) analysis was conducted to quantify the relationship between these different datasets (physicochemical parameters and WQI).
The WQI expresses chemical pollution which is strongly related to chemical ions and parameters like pH and EC. The calculation of the WQI included 11 important parameters (pH, EC, DO, HCO À 3 , Ca 2þ , Mg 2þ , Cl À , NO À 3 , NO À 2 , Fe 2þ , SO 2À 4 ), as well as their WHO standards (WHO ). The temperature is not taken into account in the calculation of WQI because the presence of ions is not related to the temperature. The WQI formula does not take into account the contribution of CO 2À 3 . EC and TDS both express the conductivity of a water sample. The two parameters are almost identical. Due to the fact that no standard is defined for TDS, EC is used in this study to represent the conductivity.
The WQI was calculated using the following formula: where q n : quality of n th water quality parameter W n : Unit weight of n th water quality parameter S n : permisible value of n th water quality parameter V n : Estimated value of n th water quality parameter at a given sample location V id : ideal value for n th parameter in pure water (0 for all other parameters except the pH and dissolved oxygen, respectively, 7 and 14.6 mg L -1 ) k: Constant of proportionality.

In situ parameters
The overall physico-chemical parameters of all water samples analyzed in and around Bonkoukou are shown in Water from the WT and those from the two taps (T1 and T2) were weakly mineralized with EC ranging from 99.30 to 117.50 μS cm À1 . Of the total water samples, 19 had EC values greater than 400.00 μS cm À1 , which is the WHO threshold value allowed for drinking water. Therefore, 59.38% of the analyzed water samples exceed this EC threshold. Out of these 19 samples, nine had EC values between 400.00 and 1,000.00 μS cm À1 , which is up to more than twice as high as the WHO standard. The other 10 samples showed EC values exceeding 1,000.00 μS cm À1 and had EC values of more than 5-40 times higher than the WHO threshold. Twelve out of these 19 water samples were from boreholes and seven from wells. This means that 70.59 and 58.30% of the water samples from boreholes and wells, respectively, had EC values higher than the standard. Especially the water in the samples from well W2 and boreholes D5 and D6 were heavily charged and were therefore of very poor quality.
No health-based guideline values have been proposed for TDS by the WHO. However, the flavor of drinking

Chemical quality assessment
The WHO does not have a guideline value for the alkalinity of drinking water, but the United States Environmental Protection Agency (US EPA) sets a directive of 120.00 mg L À1 not to be exceeded for HCO À 3 (US EPA ). Concentrations of HCO À 3 ranged from 48.20 to 561.20 mg L À1 in the analyzed samples. Out of the 32 water samples, 56.25% had concentrations of HCO À 3 above the US EPA standard. The number of samples with HCO À 3 concentrations between 1 and 2 times the standard is the same as that between 2 and 5 times. There were as many analyzed water samples whose HCO À 3 concentrations were more than 1-2 times the standard as those of concentrations 2-5 times the standard. Among these water samples, 11 were from boreholes and seven from wells.
The hardness of water is relatively related to the contents of calcium and magnesium metal cations. The concentrations of Ca 2þ and Mg 2þ were in the ranges of 4.00-934.00 mg L À1 and 1.20-510.00 mg L À1 , respectively.
Six of the 32 samples had concentrations of Ca 2þ ranging from 104.00 to 943.00 mg L À1 , which is above the WHO standard for concentrations of Ca 2þ of 100.00 mg L À1 . In addition to these six samples, two other samples were characterized by the contents of Mg 2þ above the WHO standard of 30.00 mg L À1 . These non-standard contents ranged from 36.00 to 510.00 mg L À1 . The majority (75.00%) of the water samples was within the hardness standard. Nevertheless, the water in some samples was very hard. Indeed, the contents of Ca 2þ and Mg 2þ were 6-10 and 11-16 times higher than the WHO standards.
The concentrations of Cl À obtained from the analyzed samples varied from 9.90 to 1,675.60 mg L À1 . In 72.00% of the samples, the Cl À concentrations were below the WHO standard of 200.00 mg L À1 . The concentrations of the samples that were out of the norm were in the range between 241.40 and 1,675.60 mg L À1 , ranging from 2-8 times higher than the standard.
The standards authorized by the WHO in drinking water for nitrate and nitrite ions are 50.00 and 0.10 mg L À1 , respectively. Concentrations of NO À 3 varied from 0 to 3,450.78 mg L À1 and concentrations of NO À 2 ranged from 0.01 to 10.29 mg L À1 . The samples from the WT and the two taps presented no or very low NO À 3 concentrations. Of the 32 samples, 22 (15 boreholes and seven wells) had values above the WHO standard for NO À 3 . Therefore, 68.75% of the analyzed samples have too high NO À 3 concentrations ranging from 50.80 to 3,450.78 mg L À1 . Eight of these 22 samples were from 1 to 2 times, eight from 2 to 5 times, two from 5 to 10 times, one up to 12 times and three samples from 60 to 70 times higher than the WHO standard.
This means that 88.24% of the 17 analyzed boreholes and 58.34% of the 12 analyzed wells were polluted with high NO À 3 concentrations, respectively. In 18 (56.25%) out of 32 samples, the NO À 3 concentrations were above normal. The non-standard concentrations ranged from 0.11 to 10.29 mg L À1 . Of these 18 samples, five had contents between 1 and 2 times, five between 2 and 5 times, four between 5 and 10 times, two between 10 and 20 times, and two between 70 and 100 times higher NO À 2 concentrations than the suggested standard. This means that 52.9% of the boreholes and 75.00% of the wells were polluted with NO À 2 . Of all samples analyzed, 43.75% (nine boreholes and five wells) showed concurrent elevated NO À 3 and NO À 2 concentrations where the WHO standard was exceeded by at least twice the standard value.

The standard indicated by WHO in drinking water for
Fe 2þ is 0.30 mg L À1 . The concentrations of Fe 2þ in all the analyzed water samples were nil except for a single sample that certified 0.04 mg L À1 a content far below this standard.
The concentrations of SO 2À 4 found in the analyzed samples ranged from 0.00 to 1,344.00 mg L À1 . Out of all samples, 12.50% had contents higher than the WHO's proposed quality reference for SO 2À 4 for drinking water, set at 250.00 mg L À1 . The concentrations of the samples were almost 2-6 times higher than the standard. The water from the WT, the taps and those sampled in the hamlets were exempt of any SO 2À 4 . The other water resources used by these populations had variable contents of SO 2À 4 , but all were below the WHO standard.

Bacteriological quality assessment
The results of the bacteriological analysis (Table 2) show that, with the exception of one borehole with a total germ count (50 CFU mL À1 ) less than the 100 CFU mL À1 WHO standard, all water samples were characterized by a number of total germ colonies above the WHO guideline.
Although total germs were detected in the samples, a contamination with E. coli could not be proven. However, this result did not completely reject the possibility of contamination of samples by fecal matters. Further investigation will be required in future studies to assess drinking water contamination by animal excreta.

Correlation between chemical parameters
The results of the correlation coefficients between the variables are summarized in Table 3. The coefficients in bold indicate a very good correlation.
There was an almost total linear independence (r ¼ À0. 36-0.27) between the pH and the other parameters.
The same report was observed with the DO and T. Nevertheless, the EC showed a perfect dependence on TDS (r ¼ 1). These two almost identical parameters can replace each other. The variation in EC as a function of Ca 2þ , Mg 2þ , Cl À , NO À 3 and SO 2À 4 showed an excellent correlation (r ¼ 0.96-99). The bond between EC and HCO À 3 was very good (r ¼ 0.83). The correlation between EC and NO À 2 was also significant (r ¼ 0.55). On the other hand, the concentration of Fe 2þ was independent of EC. The parameters EC, HCO À 3 , Ca 2þ , Mg 2þ , Cl À , NO À 3 and SO 2À 4 were interdependent. The degree of this dependence was more amplified in this order HCO À 3 , Cl À , SO 2À 4 , NO À 3 , Ca 2þ , Mg 2þ , EC. These parameters were mutually related, to a lesser extent (r ¼ 0.50-0.56) to NO À 2 and WQI. Noteworthy, NO À 2 and WQI have the strongest correlation (r ¼ 1).

WQI of the different waters
The WQI values of the different analyzed water samples are recorded in Table 4. WQI values ranged from 5.88 to 7,510.16. These values were particularly affected by NO À 2 , SO 2À 4 , Ca 2þ , Mg 2þ , Cl À and NO À 3 . The WQI values of the water samples indicated that 37.50% of all analyzed samples were excellent, 15.63% were good, 12.50% were poor, 3.12% were very poor and 31.25% were unfit for human consumption. Among the boreholes, excellent and good water quality was represented by 52.94% compared to 47.06% of poor, bad and unfit for human consumption. Concerning well water, 41.66% were of excellent, good and average quality and 58.34% of poor, bad and unfit for human consumption.

Physico-chemical quality of Bonkoukou's water
The pH is one of the most important parameters of water quality. It influences physical and chemical water characteristics (HC ). The pH promotes the solubility of certain substances that are harmful to water quality. The waters sampled from Bonkoukou are slightly acidic and basic.
The main adverse direct effect caused by extreme pH values (5 ! pH, pH ! 11) is an increase in skin and eye irritation (HC ). The pH of any sampled water was not in this configuration. The pH values of the samples were largely in the range recommended for drinking water. Similar Total germs >100 >100 >100 >100 >100 50 >100 >100 >100 >100 >100 >100 >100 <100 CFU mL À1 In terms of CO 2À 3 , the alkalinity of all water samples is good, but it was differently appreciated compared to HCO À 3 . The presence of HCO À 3 in more than half of the analyzed water samples shows a strong capacity of the water to neutralize acids, but it is not available for drinking in the daily life due to its high alkalinity (FEPS ).
An abnormal hardness was observed in one-quarter of the analyzed water samples, which falls in the category of 'great concern'. The water samples W2, B5, B6 are not fit for human consumption. The natural sources of this hardness are mainly sedimentary rocks, soil infiltration and stream. In general, hard water originates from areas where the arable layer is thick and the rocks are calcareous.
Some samples in Bonkoukou and the surrounding villages, in particular, water samples from the boreholes, which are most often used for drinking, contained alarming concentrations of Cl À . With concentrations above the standard, associated with the predominance of calcium and magnesium cations, chloride gives a bad taste to water NO À 2 are less present than NO À 3 . They are a result of the degradation of organic matter. In particular, nitrites come from the reduction of nitrate. NO À 3 has considerably degraded the quality of drinking water. The abnormal con- The good correlation of NO À 3 with HCO À 3 , Cl À , SO 2À 4 , Ca 2þ , Mg 2þ proves that these ions come from the same origin. In addition to the anthropogenic origin of NO À 3 , it would be important to consider and assess the contribution of minerals.
Fe 2þ was almost non-existent in these analyzed water samples, which can be explained by the slightly acidic pH values, which largely prevent the existence of Fe 2þ (HC b). Water pollution assessment using WQI The assessment of water pollution using the WQI method conducted in this study is the first of its kind in Niger.
From the correlation between WQI and the Water quality is preserved when its quality is excellent or good. A threat of deterioration is not observed for water of excellent quality and it is minor when the quality is good (CCME ). Water of excellent quality is exclusively used for drinking and water of good quality is used for domestic purposes. These two categories of water quality are also used for irrigation. Water of poor and very poor quality is almost always threatened or subject to further degradation (CCME ). Water of poor quality should only be used for irrigation. This use of water of very poor quality should be restricted. Water unfit for consumption must be properly treated before use. This is in perfect agreement with the analysis of physicochemical parameters in relation to the potability of water defined by WHO ().
Understanding the bacteriological quality water To reduce the microbial load of these samples, treatments such as filtration or boiling can be used individually or in combination. In addition, to improve the bacteriological quality, disinfection by using a disinfectant, such as hypochlorite, is recommended.  The physico-chemical and bacteriological quality of these rural groundwaters are severely affected by market gardening activities and the lack of household waste and wastewater management systems. The current consumption of these groundwaters exposes the local communities to an enormous health risk. Thus, it is imperative to prohibit the consumption of water with physico-chemical and bacteriological concentrations that are higher than the WHO standards, such as water from wells in the study area.

CONCLUSION AND RECOMMENDATIONS
In order to prevent this pollution, the best way is to raise public awareness of hygiene measures and to avoid excess application of fertilizers in agriculture. Therefore, a largescale distribution of water from the small drinking water supply system and the application of chemical disinfectants are alternative solutions for improving access to safe drinking water in rural areas, such as Bonkoukou and the surrounding villages.
In a wider context, this study also shows that, in addition to the well-known problem of access to water in dry and semi-arid regions in general, poor and deteriorating water quality can be a major problem and further endangers access to safe water.