A single methodology was adopted at two study sites. First, an analysis of the lithological profile and groundwater depth was carried out to identify the nature of the aquifer (confined or unconfined) and the type of aquifer (weathered materials or fractured layers) crossed by the equipped boreholes. Next, based on known hydrogeological characteristics, the conditions for using the analytical solution of [25] are satisfied with a reasonably estimated value set for the observation radius (≤10 m) between the piezometer and the well. The results of the hydrodynamic property values are generated automatically with the lowest associated error offset. The transmissivity value can be calculated by logarithmic approximation based on the simplification assumptions suggested by [8] according to Equation (1):

With: T: transmissivity expressed in (m²/s);

Q: pumping flow rate of the first stage at (m³/s);

Δs: variation of the drawdown (m) over one logarithmic cycle of time.

The variation of the drawdown (m) is expressed as follows (Equation (2)):

With:

S₁: drawdown at the beginning of the logarithmic cycle of time.

S₂: drawdown at the end of the logarithmic cycle of time.

4.2.1. Collection and Preparation of Hydrochemical Data

The hydrochemical data are derived from physicochemical analyses of groundwater samples collected in Guitry and Zikisso. They concern electrical conductivity and the main major ions (Ca²⁺, Mg²⁺, HCO 3 − , Cl⁻, SO 4 2 − , NO 3 − ), which are commonly used to characterize groundwater. Concentrations were expressed in mg·L⁻¹ and then converted to meq·L⁻¹ where necessary. The reliability of the data was ensured by consistency checks prior to statistical analysis, in accordance with recommendations in the literature ([12]; [2]).

4.2.2. Hydrochemical Classification according to Piper’s Diagram

The hydrochemical characterization of the waters was carried out using Piper’s diagram. This diagram, commonly used with good results ([16]; [22]), made it possible to account for the distribution of water points in the hydrochemical facies. Groundwater classification was performed using Piper’s diagram, based on the relative proportions of the main cations (Ca²⁺ and Mg²⁺) and the main anions ( HCO 3 − , Cl⁻, SO 4 2 − , NO 3 − ). This method allows hydrochemical facies to be identified and changes in the chemical composition of water along underground flow paths to be interpreted. The Piper diagram is widely used in studies of bedrock aquifers to distinguish between recent recharge, progressive mineralization, and surface influences according to ([23]; [3]; [20]).

4.2.3. Statistical Analysis and Correlation Matrix

A descriptive statistical analysis was followed by the calculation of Pearson’s correlation coefficients between electrical conductivity (EC) and major ions. This approach allows linear relationships between parameters to be identified and groups of ions that evolve together to be highlighted. High correlations between certain cations and anions are interpreted as reflecting natural mineralization processes, while associations involving nitrates are used as indicators of anthropogenic influences related to surface inputs ([2]; [1]).

Analysis of the vertical lithological profiles (Table 1) of the boreholes revealed that the aquifers in Guitry are mainly confined (80%), while those in Zikisso are exclusively confined (100%). In Guitry, one borehole (F1) crossed a semi-confined aquifer. At two sites, the aquifers are structured as multi-layered fissures (60%) or a single fissured layer (40%), which may reflect the heterogeneity or anisotropy of the aquifers encountered in Zikisso and Guitry.

Table 1. Different types of aquifers encountered.

5.2.1. Boreholes Depths

Statistical analysis of the hydraulic parameters (Table 2) of boreholes located in Guitry and Zikisso reveals significant differences in terms of depth, static level, critical flow rates, and operating yields. In Guitry, the depths of the boreholes ranged from 90 m to 110 m, with an average of 99.10 m and a standard deviation of 7.35 m, while in Zikisso, they ranged from 81.95 to 120 m, with an average of 107.74 m and a standard deviation of 17.96 m.

Static levels also show significant differences: from 1.85 to 5.45 m in Guitry, with an average of 3.62 m, compared to 2.87 to 10.82 m in Zikisso, with an average of 6.61 m. These higher levels in Zikisso suggest faster recharge and also reflect lower aquifer productivity.

Table 2. Statistical analysis of parameters.

ϭ: Standard deviation.

5.2.2. Critical Flow Rate

In terms of critical flow rates (Table 2), the Zikisso boreholes show a higher range, between 11 and 32.50 m³/h, with an average of 19.18 m³/h and a standard deviation of 9.94 m, while those in Guitry vary from 5.92 to 16 m³/h, with an average of 10.28 m³/h and a standard deviation of 4.31 m. These values reflect the improved capacity of the Zikisso aquifers to withstand high pumping rates without disrupting the hydraulic functioning of the borehole, such as wall collapse, clogging, or sand suction. In both cases, operating flow rates were set uniformly at 25 m³/h, which could prove excessive for some Guitry wells, where the critical threshold is often exceeded, thus risking compromising the sustainability of the structures if no regulation is applied.

Overall, these results reflect higher productivity in Zikisso wells compared to those in Guitry, which corresponds to the characteristics of the aquifers tapped, mainly deep fissured multilayer aquifers in Zikisso, compared to more heterogeneous structures (single layer, double layer, and multilayer) in Guitry. This qualitative and quantitative difference justifies differentiated management of the resource according to the area, with exploitation policies adapted to the fragility or robustness of the identified aquifers.

5.2.3. Hydrodynamic Parameters

The assessment of hydrodynamic parameters (Table 2), such as specific flow rate, transmissivity, and permeability, made it possible to evaluate the behavior of the aquifers exploited in the localities of Guitry and Zikisso. Specific flow rates, which are direct indicators of the productivity of the structures, vary in Guitry from 0.14 to 0.94 m²/h, with an average of 0.44 m²/h. These values highlight the productivity of the aquifers. In Zikisso, on the other hand, specific flow rates range from 0.24 to 3.27 m²/h, with an average of 1.24 m²/h, revealing productivity slightly higher than that of Guitry.

In terms of transmissivity, a key parameter defining an aquifer’s capacity to transmit water, the two areas have similar values, but are slightly more homogeneous in Zikisso. In Guitry, the values range from 1.01 × 10² to 1.05 × 10² m²/s, with an average of 1.03 × 10² m²/s and a standard deviation of 1.58, while in Zikisso, they range from 1.01 × 10² to 1.04 × 10² m²/s with an average of 1.03 × 10² m²/s and a standard deviation of 1.29. This uniformity at Zikisso reflects a better-connected reservoir geometry, probably linked to a more favorable fracture density in the bedrock, as suggested by the observations of [11] and [6] on fractured aquifers in bedrock areas.

Permeability values, derived from transmissivity and estimated aquifer thickness, also confirm this trend. In Guitry, permeability coefficients range from 10¹ to 10⁵ m/s, while in Zikisso they are slightly lower, ranging from 10¹ to 10⁴ m/s. This difference suggests better hydraulic cohesion of the aquifers in Zikisso, reinforced by a homogeneous multilayer structure of deep fissures in all the boreholes studied.

These hydrodynamic parameters confirm the greater productivity of the Zikisso aquifers compared to those of Guitry. This finding corroborates the hydraulic data and reflects the more favorable nature of the fractured bedrock at Zikisso, which offers effective permeability and greater groundwater transmission capacities. This situation suggests a more sustainable exploitation potential at Zikisso, provided that rational pumping management respects the limits defined by the critical parameters observed.

Analysis of the hydrochemical results obtained from Piper diagrams (Figure 3), principal component analysis (PCA), and the correlation matrix reveals a consistent structure in the chemistry of the Guitry and Zikisso aquifers studied. The hydrochemical facies obtained explain mineralization processes dominated by water-rock interaction, caused locally by anthropogenic surface inputs. Bedrock aquifers in West Africa are characterized by this organization, as several researchers have proven in previous studies.

Principal component analysis (PCA) was applied to reduce the size of the dataset and identify the dominant factors controlling groundwater chemistry. PCA was performed using the correlation matrix of normalized variables. The selected factor axes (F1 and F2) correspond to the components explaining the largest share of the total variance. The correlation circle and the projection of the samples onto the factorial planes made it possible to distinguish the natural mineralization processes linked to water-rock interaction and residence time from those associated with anthropogenic inputs, mainly marked by nitrates ([24]; [13]; [26]).

(a)

(b)

Figure 3. Piper’s diagram of Guitry (a) and Zikisso (b).

Analysis of the results of the Piper diagrams shows several distributions by hydrochemical classes of Guitry groundwater, while also showing a gradual evolution of mineralization. Low mineralization of Class I waters, dominated by calcium and magnesium bicarbonate ions (Ca²⁺, Mg²⁺, and HCO 3 − ), corresponds to recently recharged waters. These waters are characterized by a relatively short residence time and limited water-rock interaction. However, Classes II and III show more pronounced mineralization. A joint increase in Ca²⁺, Mg²⁺, SO 4 2 − , and NO 3 − concentrations is observed, along with a longer residence time and an increased contribution from mineral dissolution processes. A comparable study was observed in the Afema mining environment, where facies differentiation reflects the hydrochemical evolution of underground flow paths ([26]).

In Zikisso, the results from the Piper diagram show a more homogeneous distribution of samples, dominated by calcium and magnesium bicarbonate facies. This homogeneity explains the relatively uniform lithological control and the predominant role of natural mineralization processes. Studies by [20] and [15], have shown that this type of facies is typical of fractured aquifers developed in crystalline basement formations, where silicate hydrolysis is the main mechanism for dissolving major ions.

Analysis of the correlation matrix (Table 3) highlights strong relationships between electrical conductivity (EC) and several major ions, notably Cl⁻ (r = 0.79), Mg²⁺ (r = 0.76), Ca²⁺ (r = 0.68), and HCO 3 − (r = 0.66). These correlations explain why the total mineralization of water is essentially controlled by dissolved ions resulting from the interaction between water and rock. Ca²⁺ and Mg²⁺ ions have a strong correlation (r = 0.99), as do these cations and bicarbonates (r ≈ 1). These results reflect a common origin linked to the hydrolysis of silicate minerals and the dissolution of CO₂ from the soil. The work of [26] is similar to this hydrochemical behavior, attributing these ionic associations to the residence time of water in aquifers and the intensity of geochemical exchanges.

Table 3. Correlation matrix among major ions.

The results also show a strong correlation between nitrates and sulfates (r = 0.95), calcium (r = 0.91), magnesium (r = 0.91), and bicarbonates (r = 0.89). These strong correlations can be explained by the fact that they share a common origin linked to surface inputs, namely rainwater infiltration leaching soils and unsaturated horizons. Nitrates are therefore recognized as tracers of human origin. They are mainly associated with agricultural activities and the use of nitrogen fertilizers, as also shown by [1] and [26] in comparable hydrogeological contexts.

Geochemically, these processes correspond to the hydrolysis reactions of sodium and calcium feldspars, typically described by the following reactions (R.1) and (R.2):

These minerals are abundant in rock formations, such as the alteration of albite and anorthite. These theoretical reactions lead to the gradual release of alkaline earth cations (Ca²⁺, Mg²⁺, Na⁺) and the production of bicarbonate ions, while causing the neoformation of clay minerals. Although the altered minerals are not directly identified by the available data, the set of ionic associations observed, the PCA results, and the evolution of hydrochemical facies are consistent with this interpretative model, which is widely documented in the rock aquifers of West Africa ([20]; [15]; [26]).

To reinforce these results and their interpretations, Principal Component Analysis was performed. Factor F1, which includes EC, Ca²⁺, Mg²⁺, HCO 3 − , SO 4 2 − , and Cl⁻, represents the natural mineralization process linked to silicate hydrolysis and the residence time of water in the aquifer (Figure 4).

(a)

(b)

Figure 4. Community circle of the F1-F2 factorial plane of Guitry (a) and Zikisso (b).

This factor corresponds to the dominant mechanism controlling groundwater chemistry, as described in the bedrock aquifers of Côte d’Ivoire and other regions of West Africa ([19]; [26]). Factor F2 is associated with nitrates and sulfates, reflecting the influence of surface inputs and human (anthropogenic) activities on water quality.

The projection of samples in factorial plans F1 and F2 highlights a clear contradiction between highly mineralized waters, due to a longer residence time, and waters influenced by nitrate-rich surface inputs (Figure 5).

(a)

(b)

Figure 5. Factor representation of samples of Guitry (a) and Zikisso (b).

This could be explained by the dual origin of mineralization: on the one hand, a non-anthropogenic process controlled by geology and underground flow. On the other hand, a human-induced process is caused by agricultural activities and land use. The study of the Afema mining environment reached similar conclusions, where Principal Component Analysis (PCA) further identified these two major mechanisms ([26]).

In summary, the hydrochemical results obtained show that the chemistry of the groundwater in Guitry and Zikisso is mainly governed by the hydrolysis of silicate minerals and the residence time of water in fractured aquifers, with a significant but secondary contribution from human inputs through surface infiltration. This hydrochemical mechanism is consistent with the conceptual models established for bedrock aquifers in West Africa and is fully in line with the interpretative framework proposed by ([26]). However, it is important to note that the small sample size in Guitry and Zikisso is a constraint imposed by the results of hydrogeophysical investigations, which led to the drilling of wells delivering a satisfactory flow rate for consumption. Consequently, these results are specific to the area studied and should not be generalized to neighboring areas without prior studies confirming the continuity of the hydrogeological context.