The Hills Department, which covers an area of 13,931 km², is located between 7˚27'' and 8˚46'' North latitude and between 1˚39'' and 2˚44'' East longitude. Rainfall and temperature are two factors that influence the availability of water in hill towns. The rainfall regime follows a bimodal distribution in the South and unimodal distribution in the North (Bokonon-Ganta, 1987; Boko, 1988; Afouda, 1990; Houssou, 1998; Ogouwalé, 2006). And despite the amount of water flowing and seeping in, drinking water supply remains problematic due to the geological nature of the subsoil.

Geologically, the department of hills is located on the scale of the structural unit of the plain of Benin and presents a great complexity of geological formations affected by several phases of deformation, metamorphism and magmatism (Boukari, 1982). Figure 1 shows outcropping lithological formations. In the southern part, migmatitic and gneissic formations are exposed, intruded by granitic plutons in circumscribed masses and a volcano-sedimentary series (Breda, 1989; Adissin, 2012). These outcrops have a general north-south direction (Dubroeucq, 1967). In the Northeast, we observe in places, plutons of porphyroid granite with biotite appearing in the form of plissote massifs placed in the host gneissic to migmatitic, the North-West being dominated by migmatites and gneiss of the Pira group.

Studies (Kamagaté, 2006; Kamagaté et al., 2007; El-Fahem, 2008; Kamagaté et al., 2008) on the hydrogeology of the basement aquifers of Benin, particularly in the upper valley of the Ouémé and on the characteristics of the aquifers in the study area, particularly in its southern part in Dassa-Zoumé and its surroundings (Boukari, 1982; 2007) have shown the existence of two main reservoirs,

superimposed and in permanent contact, which characterize the study area. According to the authors, these aquifer systems are made of a superficial alterite reservoir made up of the semi-permeable alteration product of the underlying healthy rock, characterized by a good capacitive function and whose water table is fed from the surface by rainfall recharge; and an underlying reservoir of cracks and fractures allowing drainage of the upper loose cover with interstitial porosity.

The assessment of the physico-chemical quality of groundwater in the hills was possible using water sampling taken during a campaign in October 2019. Thus, the physico-chemical parameters measured in situ and at laboratory, were used to assess the quality of this drinking water.

• Sampling

Sampling was done at 21 water points which are boreholes intended for the drinking water supply (DWS). Figure 2 shows the spatial distribution of the sampling points. The size of the sample covers the south of the department widely. To do this, 1.5 liter plastic bottles previously washed and rinsed in the laboratory were completely filled with water. These are then hermetically sealed to prevent any gas leakage. At the same time, a 250 ml glass beaker rinsed with

the water to be analyzed (before taking measurements) and with distilled water (before any new sampling) was used for in situ measurements of pH, conductivity and TDS. Once the samples have been taken, the vials of bottles are placed in a cooler containing ice to be stored at a temperature of 4˚C until the Central Laboratory for Water Analysis (LCAE) of the Directorate General for Water (DG-Eau); where they are kept at the same temperature until the date of analysis.

• Assay methods

The pH, the temperature (T˚C), the total dissolved salt (TDS) and the Electrical Conductivity (CE) were measured by a pH/Oxi meter WTW 340i whose probe rinsed and immersed in a beaker containing the sample displays on the screen and after stabilization, the relative values of the three parameters in situ.

Chemical parameters such as calcium (Ca²⁺), magnesium (Mg²⁺) and chloride (Cl⁻) are measured by the volumetric method. Other parameters including color, sulfate ( SO 4 2 − ), nitrite ( NO 2 − ), nitrate ( NO 3 − ), phosphate ( PO 4 3 − ), fluoride (F⁻), iodide (I⁻), Ammonium ( NH 4 + ) and iron (Fe²⁺) are made by the method of spectrometry. The assay protocols for these parameters were made by standard analysis methods as described by Rodier et al., (2009) or according to the catalogs of the equipment used.

The statistical study carried out by ACPN gives numerous results which are presented in Table 1 and Table 2. In Table 1, are recorded the eigenvalues, the variances expressed for each factor and their accumulations. The factor F1, with an expressed variance of 47.58% is the most important of all, then the factors F2, F3, F4 and F5 with respectively 13.06%; 10.33%; 7.36%; and 6.07% of the expressed variance.

These five factors reflect most of the information sought and make it possible to represent the cloud of points in a significant way because the sum of the variance expressed by these factors is 84.40%. The contribution of the different variables in the definition of the main factors is given in Table 2. Each factor is defined by a certain number of essential variables in the demonstration of the mechanism of water mineralization. This table shows that the factor F1, the most important is defined by the electrical conductivity, the hardness, the alkalinity and the ions Cl⁻, Mg²⁺, Ca²⁺, HCO 3 − , NO 3 − , SO 4 2 − in opposition to the ions Fe²⁺, NH 4 + , NO 2 − and I⁻.

Significant links existing between the different parameters are given by the correlation matrix. These links reflect the different correlations that exist between the parameters analyzed. Based on the critical correlation coefficient r = 0.64 (Mangin, 1970), electrical conductivity is strongly correlated with the majority of ions (r > 0.5). Note that electrical conductivity (EC) describes the inorganic salts present in solution in water. The space of the variables of the factorial plane F1 - F2 (Figure 3) shows that this plane expresses 60.64% of the expressed variance. Factor F1 (47.58%) is determined by EC, hardness, alkalinity and Cl⁻, Mg²⁺, Ca²⁺, HCO 3 − , NO 3 − , SO 4 2 − ions, as opposed to Fe²⁺, NH 4 + , NO 2 − and I⁻ ions. The elements which define this factor come from a long duration of solution following the water-rock contact. These elements come from the hydrolysis of minerals present in the rocks which constitute the bedrock of the aquifers which shelter the waters of the region. Indeed, hydrolysis being a slow process, factor F1 therefore expresses the phenomenon of mineralization-residence time. The grouping of the majority of the variables supported by mineralization, around this axis shows the influence of alteration-hydrolysis in the dissolution of ions. The presence of NO 3 − on this axis reveals that the mineralization of water is accompanied by anthropogenic pollution.

The opposite ions of Fe²⁺, NH 4 + , NO 2 − et I⁻ drain the pole of colored waters rich in Fe²⁺, NH 4 + , NO 2 − et I⁻. Factors F2 and F3 respectively explain 13.06% and 10.33% of the inertia of the cloud of representative points of the structures and are determined by the temperature, the pH and the PO 4 3 − and F⁻ ions. Nitrogen compounds come from the degradation of organic matter by microorganisms in the surface layers of the soil, with the production of CO₂ which is then carried away in depth with the seepage water (Ahoussi et al., 2008). Better still, the naturally occurring NO 3 − ions which are part of the nitrogen cycle represent the most soluble form of nitrogen. Mainly used as inorganic fertilizers for plant growth and the synthesis of organic nitrogen compounds, excess nitrates can be found quickly in groundwater. Waste containing organic nitrogen also represents a source of nitrates obtained from various biochemical processes (ammonification and nitrification) (Aghzar et al., 2001; Amadou and al., 2014). This plan highlights the superficial exchanges that take place between the site’s water and runoff from precipitation and soil drainage.

The graphic representation in space of the statistical units of the factorial plane F1 - F2 and especially F1 - F3 highlights three main groupings of water points (Figure 4). Class 1 gathers waters of medium to high electrical conductivity, the ionic acquisition of which is under the control of mineralization-residence time accompanied by anthropogenic pollution. Class 2 contains colored waters with high and natural concentrations of Fe²⁺, NH 4 + , NO 2 − and I⁻ without anthropogenic input. These are waters with low electrical conductivity. Class 3 is much more distinct on the statistical units of the factorial plane F1 - F3. This class includes much more fluorinated and phosphated waters.

The Dendrogram (Figure 5) from the Ascending Hierarchical Classification (CAH) highlights three main groupings of variables. The 1st grouping consists

of CE, TDS and color. This group reports on the mineralization-residence time. The second group consists of chlorides and hardness. The 3rd group includes alkalinity, HCO 3 − , Ca²⁺, NO 3 − , PO 4 − , SO 4 2 − , Mg²⁺, F⁻, Fe²⁺, NH 4 + , NO 2 − , pH and the temperature indicate the hydrolysis of minerals and the contribution of human activities in the mineralization of water from the study area. This group also highlights the phenomenon of mineralization governed by the infiltration of rainwater into aquifers.

The pH of water provides information on its acidity and alkalinity. The pH of natural waters is generally between 6.6 to 7.8 (Nisbet et al., 1970) and it varies from 7.2 to 7.6 (Rodier, 1984). Usually, pH values are between 6 and 8.5 in natural waters (Chapman & Kimstach, 1996). The nature of the land crossed by the water is the natural cause, causing significant variations in pH. The analysis of these waters revealed that the pH is close to neutral, at the level of all the water points (Figure 6), the average pH values at the level of the study area were within the drinking water standards of groundwater (6.5 - 8.5 according to WHO standards).

Water temperature is an important factor in the aquatic environment as it governs almost all physical, chemical and biological reactions (Chapman & Kimstach, 1996). In the study area, the measured temperature shows almost no variation (Figure 6). The observed value is around 29.5˚. These values remain acceptable for drinking water standards.

The electrical conductivity of the water in the department of the hills varies from 56 to 1735 µS/cm for an average value of 612 µS/cm. The low values of CE are observed in the localities of Soclogbo and Torto in the South-East and in Dame while the high values are observed in the localities of Yawa and Kpakpaza (Figure 7).

The spatial distribution of magnesium in this department shows values between 1.95 and 69.06 mg/L for an average value of 22.54 mg/L. The low Mg values are observed in the localities of Soclogbo and Torto in the South-East and in Agbomadin, Dame, Missi, Logbo, Agbado and in Honnonkon while the high values are observed in the East of the localities of Yawa and Kpakpaza (Figureof Mg). Calcium varies from 4.81 to 97 mg/L for an average value of 46.03 mg/L. The low Ca values are observed in the localities of Soclogbo and Torto in the south-east and west of the localities of Yawa and Kpakpaza while the high values are clearly observable in Gbedje, north of Karre and on both sides other from Ouedeme.

It follows from variations in calcium and magnesium, a similar distribution with hardness, a function of Ca²⁺ and Mg²⁺. Then its values are between 24 and 498 mg/L with an average value of 208 mg/L. As a result, groundwater is less hard in the localities of Soclogbo and Torto in the south-east while it is distinctly hard in Gbedje, west of Ouedeme, Yawa and Kpakpaza (Figure 8).

The alkalinity in the hills department shows values between 10 and 80 mg/L for an average value of 43.81 mg/L. The low values of alkalinity are also observed in the localities of Soclogbo and Torto in the south-east while the high values are observed in the localities of Yawa and Kpakpaza (Figure 9).

The spatial distribution of chlorides in the department shows values that are between 14.2 and 335 mg/L for an average value of 134.89 mg/L. The low values of Cl⁻ are observed in the localities of Soclogbo and Torto in the south-east and in Dame, Missi, Logbo, Agbado and in Honnonkon while the high values are in Gbedje, Agbomadin and in Karre (Figure 10). With regard to fluoride ions, they vary from 0 to 1.1 mg/L for an average value of 0.49 mg/L. Non-flowered waters are found in the localities of Soclogbo and Torto in the south-east and west of the localities of Yawa and Kpakpaza while those strongly fluorinated are clearly observed in the south of the localities of Atchakpa and Kpakpaza (Figure 10). It should also be noted that these values are lower than the WHO guide value which is set at 1.5 mg/L.

For iodide ions, its values are between 0.09 and 2.09 mg/L with an average value of 0.39 mg/L. The iodides are strongly observed in the localities of Soclogbo and Torto in the south-east and north-west of Ouedeme while its values are very low in the north and west in the department (Figure 11).

The spatial distribution of iron in the waters of the Collines department shows values which are between 0 and 1.25 mg/L for an average value of 0.28 mg/L. Low Fe values are observed in the west and in the center of the department. High iron values are observed north of Honnoukon precisely between the localities of Agbomadin and Dame on the one hand and those of Yawa and Kpakpaza on the other hand (Figure 11).

The bicarbonates in the waters of the department of the hills vary from 6.1 to 48.8 mg/L with an average value of 26.72 mg/L. The low values of HCO 3 − are observed in the localities of Soclogbo and Torto in the south-east while the high values are in the south-east of the localities of Yawa and Kpakpaza (Figure 12). The spatial distribution of ammonium in the department’s groundwater shows values between 0 and 2.12 mg/L with an average value of 0.24 mg/L. NH 4 + values are high in the localities of Soclogbo and Torto in the south-east and almost zero in the localities of Gbedje and Lozin in the west of the study area and in the

center in the localities of Yawa, Kpakpaza, Karre, Ayedero, Zongo and Ouedeme (Figure 12).

Like ammonium, the spatial repair of nitrites is similar to it. Nitrites in the waters of the department are generally low with slightly higher values in the south-east, in the localities of Soclogbo and Torto. The values are between 0 and 0.27 mg/L for an average value of 0.03 mg/L (Figure 12). Unlike the two previous ions, nitrates are heterogeneous and vary between 0 and 163.91 mg/L for an average value of 53.07 mg/L. The values of NO 3 − are low in the South-East in the localities of Soclogbo and Torto or even Yawa, Honnoukon, Ouedeme, Agbomadin, Dame and Atchakpa. They are very high, exceeding the standard (50 mg/L) accepted by the WHO in the localities of Gbedje, Lozin, Missi and Logbo in the West, Aizon in the North-East, Karre and Zongo in the South of the zone d. study (Figure 13).

Phosphates in the waters of the Collines department are homogeneous and low throughout the department, with slightly higher values at the east of Ouedeme. The values are between 0.09 and 2.82 mg/L for an average value of 0.40 mg/L (Figure 14). Regarding the sulfate ions, they vary from 0 to 128 mg/L for a mean value of 25.9 mg/L. Waters very low in SO 4 2 − are found in the localities of Soclogbo, Atchakpa and Torto in the south-east, while those strongly sulphated are clearly observed in the South in the localities of Yawa, Kpakpaza, Karre, Ayedero and Zongo (Figure 14).

The quality of groundwater intended for consumption in the Hills was assessed following the study of the pollution parameters and the interpretation of a simplified grid (Table 3) namely the electrical conductivity and the chloride ions

which provide information on the mineralogical quality of water then nitrates which are indicators of groundwater pollution (Nouayti et al., 2015).

Analysis of the samples has shown that the water quality is generally good or even excellent (Table 4). However, a cross-reading shows very poor-quality waters (zongo, toui and Aizon) of poor quality (karré, Missi, Logbo, Lozin, Ouédemè and Gbedjè) given the nitrate contents.