INTRODUCTION
Several efforts in the world are made by the public services to manage the urban, periurban and rural centers in quality water through several programs from national and international partnerships. However, difficulties exist since portable water is still unavailable for everyone. The supply of portable water is thus oriented towards the research for underground water, whose qualities generally fulfill the international norms of the OMS (Faillat, 1986; Biemi, 1992; AbeSombo et al., 2011). The necessity to urgently and immediately satisfy the need for water of the populations in regions with hydric deficits in quality water leads to a precipitation in the realization of borings. Generally, these works realized on the bases of geomorphological considerations are no longer useable in long term due to their frequent dry up. Thus contributing to less water supply and consequently, scarcity in potable water in cities and villages especially during drought and hence a waste in finance.
The aim of this work is to understand the electrical and mechanical behavior of the aquifers of piedmont, all this by specifying the underground structure of aquifers and to determine the parameters susceptible to influence the productivity of borings, with a view to generalize this information for the guidance of hydraulic campaigns in similar regions. To this effect, the use of vertical electrical sounding data from the application of the electrical method of Schlumberger (Atangana, 1992; Robain et al., 1996; Teikeu et al., 2012a; Nouck et al., 2013), combined with data from drilling parameters (lithological logs, total drilled depth, thickness of weathered level, static level, depth of water inflow, flow), appear to be very interesting for hydrogeological characterization of aquifers in this context.
MATERIALS AND METHODS
Geologic and geographic settings of the study area: The study area also called the piedmont area is located between the mountainous area of the Mandara Mountains and the vast plains of the Logone and Chari field (Detay, 1987). It is locally dominated by Quaternary alluvial mantles named ncomblement Tchadieno meaning Tchadian fill up and ancient Tchadian continental terminal sands.

Fig. 1: 
Geologic map of the study region 
They are detrital arenas and less evolved lithosols which mask the original geologic formations. Petrographic studies in the area revealed a Precambrian crystalline basement, made on one hand by original metamorphic formations of epizonal formation (metavolcanites, tufaceous schists and dominant quartzites); non migmatitic mesozonal formations (leptinitic gneiss and quartzite); migmatitic mesozonal formations (embrechiticgneiss, anatexis granite) and on the other hand, by formations of migmatitic affinity (calcoalkaline granites) and plutonic formations, twomica (biotite and muscovite) and garnet granites which outcrop as domes in some places (Fig. 1).
This region is located in the FarNorth of Cameroon between latitudes N10°9' and 11°6' and longitudes E13°56' and 14°48' called Diamaré, the region occupies a surface area of about 4818 km^{2}. It has a dry tropical climate, characterized by high average annual temperatures (27.7°C) and two well distinct seasons. The average annual rainfall varies between 1000 mm in the Mandara mountains to 500 mm in the "Yaérés" plains. The registered absolute minimal temperature is 11.4°C against an absolute maximal temperature of 45.5°C (Ngoupayou, 1997). These climatic parameters are actually influenced by climate change phenomena and contribute to high water evaporation, drying up of streams and soils, desert advancing, leading to water scarcity in certain periods of the year (Zakari et al., 2014).
Data and methodology: Geophysics is an appreciable instrument in the guide for hydrogeological research programs, whether in the identification of new resources or in the amelioration of knowledge concerning a reservoir aquifer. The electrical method, through the Schlumberger lateral (electrical dragging) and vertical (electrical sounding) investigation techniques were applied in this study. This is because it seems to be the most adapted, due to the fact that it is light, its high credibility and favorable geological setting consists of lithologic formations having different resistivity priori (Lachassangne et al., 1996). This method is based on the injection of electric current into the ground and intensity (I) at the surface, at mid distance between the two emission electrodes (A and B). This dispositive is followed by the measurement of the potential difference (V), created by current flow through the crossed formations at the middle of the reception electrodes (M and N). This is aimed to establish apparent resistivity variation curves (ρ_{a}) with respect to the distance AB/2 (Teikeu et al., 2012b; Nouck et al., 2013).
According to Telford et al. (1990), for a homogeneous field and isotope:
This Eq. 2 allows us to deduce the true resistivity:
With:
The factor K given by Eq. 3 is called “Geometrical factor” and depends only on the relative position of the four electrodes. It is calculated from the lengths in meters.
If the field is not homogeneous, obtained from K, ΔV and I apparent resistivity ρ_{a} linked by complex relationships with true resistivities and thicknesses of all layers influenced by the AMNB measuring device.
It consists of revealing the underground geologic discontinuities that are potentially associated to water transfer (Meli'i et al., 2011; Nkoungou et al., 2012). According to Seladji et al. (2007) the measured resistivity decreases when water content and degree of saturation increases. The water content and electric resistivity ratio constitutes a major advantage for electrical resistivity methods in the search for underground waters. Darcy's law assumes a linear relationship between flow and pressure drop through a porous medium and show that, the load can be seen as the potential of the flow velocity field at each point M = (x, y, z) of the space greater than that of gravel or crack scales.
The porosity P of a porous medium is given in Eq. 4:
The specific surface area S is given by Eq. 5:
From the Eq. 5 the effective porosity n_{e} can be obtained with Eq. 6:
The storage capacity Sc for a porous medium can be expressed by Eq. 7:
In the porous medium, the gradient load or hydraulic gradient can be expressed by Eq. 8:
where, Δh is loss load between the top and bottom of the porous medium and L, the thickness of the porous medium.
According to Darcy law, the flow rate through the porous medium Q is given by Eq. 9:
where, K is the permeability and A the normal section of this porous medium.
Hydrogeological characterization of aquifers also requires the implementation of a tool whose results are complementary to those of the electrical method. It is in this context that a statistical analysis to assess the influence of parameters on the flow rates of drill works was used. For this, the correlation coefficients (r) (Pearson, 1895) between flow (Q) and various mechanical parameters (y) from the covariance between Q and y (σ_{Qy}) and standard deviations of Q and (σ_{y}) were calculated by applying Eq. 1013:
with:
and:
where, N is the total number of wells whose flow rates were measured, m_{Q} and m_{y} are averages Q and y.
All these analyses highlight the links or absence of links between the evolutions of the different parameters.
In this study, according to Eq. 1, 2 and 3, 56 vertical electrical sounding (SEV) of the Schlumberger type, were collected in the field using the Syscal IRIS Instrument electrical resistivity device with the wire lengths ranged from 1.5 120 m to the distance AB/2 and 0.5, 1.5, 2.5, 5.0, 10.0 and 15.0 m for the different distances of MN/2. Also, The Géo.Elect.Mod software was used in the interpretation of electrical sounding from Gosh linear filters combined with parameters of 68 wells (boreholes) and their drill (Fig. 2).

Fig. 2: 
Map of study area showing electrical drills and well locations 
RESULTS AND DISCUSION
Sounding curves: According to the data collected during the framework and using Eq. 2, the sounding curves was summarized into seven categories (Table 1), inspired from those realized by Koussoube et al. (2003), AbeSombo et al. (2011), Teikeu et al. (2012a) and Nouck et al. (2013).
Curves of H, HA and QH type: These curves present the form of a «boat base» with symmetric (Fig. 3a) or asymmetric (Fig. 3b, c) branches. It does not give any indication that can clarify the presence of fractures. These fractures are actually present but are often masked at the base of alterites or are mistaken for basements due to itssmall thickness at greater depths (Koussoube et al., 2003). The interpretation of these curves (Table 2), shows a structure with three geoelectrically distinct layered terrains with respect to the behaviour of the curve’s branches. The descending branch characterizes a resistant autochthonous clayey sand lateritic layer (in the basement zone) or allochthonous sandy clay layer (in the alluvial or sandy zone). Its resistance is due to climatic conditions and represents the superficial Arable recover with a relatively small thickness of 0.53 m. The conducting aquifer complex with an average thickness of 1525 m is represented by the boat base like point and the starting points of the ascending walls and is associated to the residual clayey sand weathering or sandy strata of an alluvial clayey episode. Its resistivity is relatively weaker than that of the upper layer due to its high water content, covering a confined aquifer in the lateritic zone or the unconfined aquifer (phreatic nap) in the alluvial zone. The ascending branch is characteristic to the laterite’s parent rock or a resistant colluvial arena (Fig. 3d).
Curves of KH type: With this type of curves, four terrains can be distinguished (Table 2). They have “Bell and boat baselike” forms (Fig. 4a) and are characteristic to zones where cuirassed or ferruginous gravels were formed, just after the autochthonous clayey or allochthonous silty sand lateritic superficial cover. On the profile, it is the second layer of the terrain and is more resistant than the first layer, thus giving the curve a “Belllike” shape. The other terrains are the same as those defined by a “Boat base like” form. On the hydrogeological point of view, they signal a single localizable confined aquifer in the alterite or in the weathering front of the fissured/fractured parent rock (Fig. 4b).
Curves of HH type: They present a curve with a «two consecutive boat baselike» forms (Fig. 5a) and characterize terrain models with five layers (Fig. 5b).
Table 1: 
Station location and curve types 


Fig. 3(ad): 
Geoelectrical model and general shapes of type H, AH and QH curves 

Fig. 4(ab): 
Geoelectrical model and general shape of KH type curves 
Their interpretation (Table 2) reveals two superposed aquifer levels: The first superficial aquifer is situated in the lateritic or alluvial superficial formation and is supplied by meteoric water and the infiltration of the neighbouring river “Mayo”; the second deep aquifer in the fissured basement and weathered or weathering front and supplied through underground fractures that are connected to the drainage network.
Curves of A and Q types: These are curves with a single descending branch (Q type) (Fig. 6a) or a ascending branch (A type) (Fig. 6c).
Table 2: 
Geoelectrical parameters of VES stations 


Fig. 5(ab): 
Geoelectrical model and general shape of HH type curves 

Fig. 6(ad): 
General shape of A and Q type curves 
Table 3: 
Descriptive statistics of drilling parameters 

Pt: Total depth of drilling EpAlt: Weathering thickness, EpAltsat: thickness of saturated weathered or alluvium layer Pve: Depth of water arrival, EpRocksat: thickness of saturated weathered Q_{est}: flow rate 
Table 4: 
Results of correlation coefficient 

They characterize terrain models with two or three layers when they are in a regular evolution and models with more terrains when they present irregularities in their evolution «drawling shape» (Table 2) (Fig. 6b, d).
In the case of type A curves, the drawling shape of the ascending slope is due to the presence of a very interesting level in hydrogeology which is made up of the fissured basement between the weathering front and the fresh basement. This curve denotes the influence of a mega fracture and presents perturbations at the level of the ascending branch. These perturbations correspond to interesting hydrogeological indices on the fractured or weathered zone of the underlying basement.
In the case of type Q curves, their drawling shape is due to the presence of a very water saturated and impermeable clayey level. The nonascending of the curve is due to the distance of the basement which has not been attained. The profile is made up only of alluvial formations and the aquifer is very seasonal when present.
Aquifers typology: It emerges from the analysis and interpretation of electrical sounding curves figures (Fig. 36), two types of aquifers in this geological study area: the shallow aquifer located in the regolith and alluvium due to the presence of an impermeable subadjacent layer of clay with the absorbent properties that stores water, consists primarily of meteoric waters and the contributions of neighboring seasonal rivers. Although covered in some places by the cellular ironstone facies or gravelly which protects it from harsh weather conditions that influence its rechargeability, this aquifer is seasonal; the confined aquifer is developed in the gaps left by the different tectonic events which control and direct the flow of groundwater in the first 15 m of the cracked/fractured rock. This aquifer is fed on the one hand, a part of the waters of the unconfined aquifer with whom he communicates when it is existed at the top and on the other hand, through fractures connected to the drainage network rivers.
Statistical analysis and interpretation of drilling parameters based on flow rate: The flow rates obtain with the Eq. 6 and the mechanical parameters of wells varies in the general studied population (Table 3). Generally, 32 or 47% of these wells have an estimated flow rates superior to 2 m^{3} h^{1} which according to Savane et al. (1997) is considered to be significant. About 19.12% of the flow rates are superior to 5 m^{3} h^{1} and is qualified as a strong flow by the InterState Committee for Hydraulic Study (CIEH).
The result of correlation coefficient Eq. 10 of the different parameters individually collected with the flow rates (Table 4), show the absence of a significant relation. This is indicated by low correlation coefficient values. Nevertheless, tendencies can be cleared under certain conditions.
Total depth of drillingflow rate relation: The relationship between the total depth of drilling and flow rate is given by the Eq. 14:
Y = 6.79E^{1}X^{0.149} 
(14) 
For most wells, the total depth of the drill works is situated between 50 and 80 m (Fig. 7a). In this interval, more important flow rates than low flow rates are as well found. The strongest flow rates (5 m^{3} h^{1}) do not necessarily correspond to the highest depths. The essentials of this flow rate is situated between 32 and 65 m. However, all wells with depth beyond 75 m have flow rates lower than 2 m^{3} h^{1}, this seems to be an indication that, if a significant flow rate is not found beyond the depth of 75 m, it is not necessary to continue drilling. In effect, beyond a certain depth, fractures become less dense, others are closed, thus reducing the chances of meeting important flows (Lachassangne and Wyns, 2005). To this fact, Dibi et al. (2004), Ngo et al. (2010) and Koita et al. (2010) support the idea that, in basement contexts, the most productive wells are situated between 40 and 60 m and this depth is considered as the optimal. Generally, the maximum bored depth during hydraulic programs is agreed upon from the start of the project, as well as a positively judged flow. When this positive flow is attained, drilling is immediately stopped without considering the possibility of finding more consequential flows at depth. On the other hand, some wells can be prolonged in the search for an imposed flow.
This explains the existence of very deep wells, due to the absence of consequential flows in the first meters. The correlation coefficient is 0.58, thus, leading to the conclusion which states that, there is no scientific consideration in the choice of well depth.
Weathering thicknessflow rate relation: The relationship between weathering thickness and flow rate is given by the Eq. 15:
Y = 3.26E^{3}X^{2}+2.30E^{1}X4.52E^{2} 
(15) 
It is difficult to clear a tendency, considering that the correlation coefficient is 0.32. If important flow rates are obtained between 1848 m, most wells of low flow rates will equally be found in this range (Fig. 7b). At depths less than 18 m, wells of 3.5 m^{3} h^{1} flow rates do not exist.

Fig. 7(ae): 
Well parameter variations with respect to flow 
This tendency seems to tie with that verified by Koita et al. (2010) and Kouassi et al. (2012) whereby the thickness of weathered levels or alluviums increases the possibility of obtaining important flow rates. However, they precised that, this relation is not linear since it depend on the geotechnical properties of the weathered layer. As such, once the weathered layer is sufficiently thick, more than 20 m, the productivity of the well will depend on the water content of this weathering layer, its capacity to liberate its water content (porosity) and the capacity of the underlying fractures to drain the reservoir (permeability).
Depth of water arrivalflow rate relation: The relationship between the depth of water arrival and flow rate is given with the Eq. 16:
Y = 8.40E^{4}X^{3}4.54E^{2}X^{2}+6.37E^{1}+5E^{1} 
(16) 
The distribution of points on this graphs (Fig. 7c) equally have the tendency of revealing the two types of reservoir. These reservoirs are: The superficial aquifer between 3 and 15 m, in the weathered layer and alluviums and the deep reservoir of the fissured/fractured parent rock or the underlying basement of more than 15 m. Considering water arrivals beyond 15 m to be coming from the fissured/fractured reservoir, it is revealed that, almost all flow rates above 2 m^{3} h^{1} considered by Savane et al. (1997) as significant, are obtained from the first 15 m of the fractured rock. This tie with the earlier verification above which states that, at a certain depth, fractures became less dense and closes up, thus reducing flows in the deep fractures of the basement (Lachassangne and Wyns, 2005). Generally, the correlation coefficient is 0.16 and 0.17 in the saturated fractured rock.
Relation between the thickness of saturated weathered or alluvium layerflow rate and thickness of fractured rockflow rate: The relationship between the thickness of saturated weathered or alluvium layer and flow rate is given in the Eq. 17:
Y = 3.13 E^{5}X^{3}8.80E^{3} X^{2}+3.77E^{1}X 
(17) 
and that between the thickness of fractured rock and flow rate in Eq. 18:
Y = 1.57 E^{3 }X^{3}6.81E^{2} X^{2}+7.78E^{1}X 
(18) 
From the 68 studied well, 70% captures the superficial tablecloth of the weathered or alluvium layer which presents variable thicknesses between 0.8 and 45 m.

Fig. 8: 
Hydrogeoelectric model realized from the interpretation of electrical sounding results of piedmont region in diamare 
The graphs representing the thickness of the saturated, weathered or alluvium layer with respect to flow rates (Fig. 7d), have a correlation coefficient sensibly equal to that of complete weathering (0.31). They present flow rates which increases between the thicknesses of 135 m and decrease for higher thicknesses. This can be due to the irregular distribution of porosity in the weathered layer or alluviums, in relation with the different phenomena of weathering and sedimentation which prevailed in the region.
It is important to note that, water arrival in wells is not only observed at the level of the saturated alterite but can also be observed in the part of the fissured/fractured parent rock. The graph (Fig. 7e) representing the depth of water saturated fractured rock with respect to flow rates, show that, the flow rate increases between 2.5 and 8 m and decreases for depths above 13.5 m. In some drill works, several water arrivals are felt in the same well, thus confirming the hydrogeoelectric models of two superposed aquifer as obtained through the interpretation of electrical sounding curves of the region and those from the works of Koussoube et al. (2003), Teikeu et al. (2012b) and Nouck et al. (2013).
Aquifers rechargedischarge: The Fig. 8 shows that, during rainy seasons, aquifers around the Mayos, recharge not only by rainwater through the pores and fissured rocks but also by the waters of transgression from the Mayos. However, in the dry season, under the effect of gravity and capillary aquifers lend to the Mayos, the stored water. Furthermore, this figure confirms the presence of the water, by breaking rocks in some quarries of the NorthEast region.
CONCLUSION
The present geophysical and statistical study was carried out in the FarNorth of Cameroon, precisely in the piedmont region of Diamare. It had as aim to precise and understand the structures and functioning of aquifers and to find parameters susceptible of influencing the productivity of wells. To attain this objective, the Schlumberger sloping electrical resistivity method and statistical analyses of well parameters, were the key methods employed. As such, this study has permitted firstly from the shapes of the sounding curves:
•  To identify two types of hydrogeological models, A hydrogeological model with one superficial or deep aquifer and a hydrogeological model with two superposed superficial and deep aquifer 
• 
To localize a superficial aquifer in the weathered layer or in sandy strata of clayey alluvial episodes issued by the accumulation of eroded materials from the Mandara mountains, by the action of intense fluviotorrential erosion and a deep aquifer in the fissured and fractured zones of the parent rock created by tectonic activity and connected to the drainage network 
• 
To propose by intelligent modeling, an hydrogeoelectrical model of the piedmont region through the shapes of the characteristic sounding curves of soils developed in the tropical zone, where the phenomena of weathering, erosion, sedimentation and pedogenesis have printed their marks 
Secondly, from the examination of graphs from the characteristic well parameters, individually collected with respect to the flow rates. These examinations lead to the following conclusions:
•  The depth of a drill work is not a guaranty to an important flow in the piedmont region; beyond the drilled depth of 75 m or beyond 35 m of saturated weathered or alluvium layer, the chances of meeting an important flow rates are reduced 
• 
The important water arrivals in deep aquifers are observed in the first 15 m of the rock 
• 
The productivity of drill works is positively influence by the thickness of the weathered and alluvial levels but the weathered and alluvial levels has to be water saturated and sufficiently porous and permeable to liberate its water content 
• 
The extreme variability of flows rate in the region of a borehole to another, highlighted the very discontinuous nature of these aquifers foothills 
• 
Finally, this study has enabled to understand the aquifers rechargedischarge processes in this piedmont 
ACKNOLEDGMENT
The authors thank Mrs Akozo Marie Laurence and the late Mr. Jeannot Nouck and Mrs. Julienne Ngo Matip for the financial support, assistance, general comments and suggestions. They also gratefully acknowledge Sundus Mariam and the anonymous reviewers for their detailed comments that have considerably improved this paper through their pertinent and constructive remarks.