Introduction

Water is a vital resource for human, animal and plant species. It is important for the existence of all living beings, the survival of the environment as well as the blooming of industry and businesses. Water is also very important for agricultural activities. Unfortunately, in most parts of the world, water is a scarce resource. Hence, groundwater is an important part of the policy of water resources management in many countries. This is the case of Cameroon where a series of investigations are focussed on the Adamawa region, nicknamed “water tower”. This region gave room to a lot of investigations; the outcomes are to be revised, updated and improved in the present article. The interest observed nowadays in groundwater resources is due among others to the scarcity of surface water resources or their poor quality (Aretouyap et al. 2020; Stelmach and Clasen 2015).

For a lasting and sustainable exploitation of groundwater resource, geoelectrical methods are of paramount importance as they allow assessment of hydrogeological parameters, such as transmissivity and hydraulic conductivity, as well as flow from transverse resistance, longitudinal conductance, etc. In fact, strong relationships are possible between electrical data measured by geoelectrical methods and hydrogeological parameters. Lesmes and Friedman (2005) established several relationships between the hydrogeological and the electrical properties of soils and rocks. From basic electrical data, such as low-frequency conductivity and high-frequency permittivity, these relationships are useful for lithological characterization, saturation monitoring, salinity prediction and assessment of the porosity of fully saturated geological formations as well as the water content of the partially saturated geological formations. Inversely, some electrical data such as the electrical conductivity can be predicted from petrophysical properties. Mawer et al. (2015) built a numerical model from electrical properties to generate hydrogeological ones. Then, they established a useful power exponent equation to compute the aquifer hydraulic conductivity. The resulting prediction was very accurate when the electrodes are properly arranged with regard to the presence of clay. Kirkby and Heinson (2015) also modelled and used random resistor networks to link aquifer hydraulic and electrical properties. Singh and Singh (2016) combined electrical parameters with pumping test data and deduced the hydrogeological characteristics of Tuticorin aquifers. They determined the true thickness and resistivity by interpreting the data measured in the neighbourhood of the pumping test location. Finally, several empirical relationships were established and used to estimate the hydrogeological parameters from the electrical data. Tijani et al. (2018) estimated the hydraulic properties of basement aquifer from resistivity sounding data in Ibadan. In the same way, Youssef (2020) characterised and assessed the aquifer hydraulic properties in Ain El-Soukhana from electrical data. More recently, De Almeida et al. (2021) used VES to estimate the hydraulic parameters of a porous aquifer at the experimental farm of the University of Brasilia.

The present investigation uses the same approaches abovementioned. Electrical data have been collected from 50 VESs conducted within the region with the purpose of deducting the hydrogeological parameters of the local aquifers. Those parameters have been determined progressively since 2014 with significant gradual improvements. The main objective is to revise, update, organize and improve various significant results gathered in the area (Aretouyap et al. 2014a, b, 2015, 2017, 2018, 2019a, b; Bisso et al. 2019).

Geophysicists have realized that a correlation between the hydrogeological parameters and geoelectrical properties of an aquifer is of paramount importance. So, selecting the correct hydro-geophysical methodology is considered as an optimal solution. Many relations between hydraulic and geoelectrical parameters of various aquifer types have been studied and reviewed by several researchers (Kelly 1977; Niwas and Singhal 1981, 1985; Mazac et al. 1985; Huntley 1986; Frohlich et al. 1996; Chen et al. 2001; Niwas and de Lima 2003; Singh and Frevert 2005; Hubbard and Rubin 2000; Soupios et al. 2007; Chandra et al. 2008; Khalil and Santos 2009; Massoud et al. 2010; Niwas et al. 2011; Sikandar and Christen 2012; Niwas and Celik 2012; Singh and Singh 2016). Such relations can be used as a helpful tool to estimate the hydraulic parameters mathematically with an acceptable degree of accuracy and allow getting the maps of transmissivity and hydraulic conductivity of the entire study area.

The region of Adamawa, because of its important role as major catchment for four riparian countries, namely Cameroon, Central African Republic, Congo and Nigeria, has been the focus of intense research in the domain of groundwater resources for several years (Aretouyap et al. 2014a, b, 2015, 2017, 2018, 2019a, b; Bisso et al. 2019). Results include the impact of climate change on local water resources, analysis of the groundwater quality, describing the local lithology and the evaluation of geoelectrical and hydrogeological parameters of the local aquifer. However, with these investigations, values of hydrogeological parameters obtained made only slight or moderate improvements. This paper aims at revising geological conception, datasets, geophysical approaches and key results found in previous works related to groundwater resources in the region. The main outcome will be the update of key groundwater parameters.

Geology and morphology

The present paper focuses on the Adamawa region because of its hydrogeological importance. As it is, 150–300 km wide, the Adamawa Plateau is called “the water tower of the region” because it feeds major watersheds in Cameroon, Central African Republic, Congo and Nigeria. This region is located between 6°–8° N and 11°–16° E in the heart of Central Africa as shown in Fig. 1. It extends 410 km from west to east between the Central African Republic and the Federal Republic of Nigeria, for a total area of 6782 km2. The morphology of the region is dominated by volcanic highlands, resulting from tectonic uplift and subsidence (Tchameni et al. 2001). The local 1100 m altitude terrain is very rugged with an uneven escapement of several hectometres and a large cliff in its northern part.

Fig. 1
figure 1

Geological map of the Adamawa Plateau (Maréchal 1976) as amended

Full size image

In the centre of the area, one can observe gentle topography and swampy valleys, scattered with mountains and volcanic cones. The eastern part is marked by huge massifs resulting from erosional and tectonic events. In the west, high mountains and hills are observed. One notes also plains and basins such as the Tikar plain. A huge plateau sloping gradually is found in the southern part of the region (Toteu et al. 2000).

The local basement is made up of the Pan-African granite-gneiss, represented by Ordovician granites, gneisses and migmatites. The main geological units present in the region are trachytes, basalts and trachyphonolites based mostly on concordant calco-alkaline granites and discordant alkaline granites (Toteu et al. 2000). There is also the presence of metadiorites of Paleoproterozoic basement in some places (Toteu et al. 2001). Faults observed in this region are grouped into two main directions:

  • “Volcanic line of Cameroon” that contains the most common direction-oriented N 30° E;

  • “Adamawa shear zone” oriented N 70° E.

Contrary to previous studies (e.g. Aretouyap et al. 2015) where the region was considered as a unique and homogeneous geological unit when determining and interpreting aquifer parameters, we subdivide the VES database into two groups corresponding each to a major geological formation: metamorphic and sedimentary as shown in Fig. 1. The granite-gneiss bedrock is represented by Ordovician granites, gneisses and Pan-African migmatites (Toteu et al. 2000). This setting results from a series of three geologic events: a long period of continental erosion from Precambrian to Cretaceous, an onset of volcanism from Cretaceous to Quaternary and recurrent basement tectonics.

Recent geophysical investigations conducted by Njonfang et al. (2008) reveal intense faults in the bedrock. As a result, the groundwater potential is important in the studied zone. This potential has been explored, with various electrical methods, by Aretouyap et al. (2015, 2017, 2018, 2019a, b) and Bisso et al. (2019). In addition to the fact that those investigations considered the whole region as a single geological unit, electrical methods and techniques used for investigation have improved gradually. As a result, the parameters determined successively have also progressively changed. This situation justifies the interest of this paper which aims at updating key parameters of groundwater in the region. More accurate hydro-parameters of the regional aquifers obtained from various recent investigations will be produced. The study area has up to 50 VESs with the same number of boreholes mostly tapping a fractured aquifer. Thirteen of these boreholes have been used for experimental pumping test operations.

Methodology

The methodology section brings together field investigations and previous works carried out within the study area to establish different relationships between electrical data and hydrogeological parameters. For this, VES data adjacent or close to experimental wells (with pumping tests results) are used to draw correlations between electrical and hydrological parameters. Then, these correlations are used to predict and interpolate the hydrological parameters from VES results which are not close to wells. For this geophysical campaign, 50 VESs in a Schlumberger configuration with the half current electrodes separation varying from 1 to 300 m were carried out. 13 of these soundings are carried out in the vicinity of experimental wells used for pumping tests. Before any interpretation, the curve-matching technique, as stated by Orellana and Mooney (1966), was used for calibration.

Establishing relationships between electrical data and hydrological parameters

Thirteen out of fifty relevant wells/boreholes randomly distributed within the study area were used for pumping tests and other experimental measurements to determine several parameters, such as transmissivity, hydraulic conductivity, depth, thickness, etc. For the VES points located nearby those experimental boreholes, electrical data were interpreted using a conventional curve-matching method (CMM), as in Zohdy (1989). The outputs of this interpretation are relationships between aquifer parameters and geoelectrical data as well as the aquifer modelling. In addition to CMM, ISM was also used. The ISM approach consists in plotting the conductance (AB/2)/ρa versus the electrode separation (AB/2). Each line segment resulting from this graph represents a layer; and the intersection of the line segments, multiplied by a factor of 2/3, corresponds to the depth of a particular layer. Thus, if t1, t2, …, tn are intersection points (along the AB/2 axis), then 2/3 × ti is equal to the depth of the ith interface (Asfahani 2016). Similarly, when (AB/2)/ρa is plotted versus AB/2, the inverse slope of the segment gives the resistivity of the ith layer except when a segment, representing the high-resistivity layer, shows a negative slope. In this case, the resistivity of the layer is taken as infinity (Asfahani 2016). This method has the ability to decipher thin layers even if they are buried at great depths.

Hence, five main parameters of the local aquifer have been computed and updated using resistivity methods. These parameters are aquifer transmissivity, hydraulic conductivity, resistivity, thickness and depth which are very useful for groundwater extraction in this region. For example, it is possible to correlate the resistivity with the hydraulic conductivity as well as the transverse resistance with the transmissivity (Kelly 1977).

The interpretation of thirteen VES points located in the vicinity of experimental wells allowed determination of the aquifer resistivity (ρe), thickness (h) and transvers resistance (TR). By measuring the water resistivity (ρw) from those wells, the formation factor (F) is calculated using Eq. 1.

$$ F = \rho_{{\text{e}}} /\rho_{{\text{w}}} . $$
(1)

The aquifer hydraulic conductivity (K) and transmissivity (T) are measured from the pumping tests. These electrical data and hydrogeological parameters are presented in Table 1.

Table 1 Hydrogeological parameters from the thirteen experimental sites
Full size table

Predicting hydrological parameters from VES data without experimental wells

From the experimental data presented in the above Table 1, hydrogeological parameters were plotted against electrical data as shown in Fig. 2 and linear empirical regressions as expressed in Eqs. (2)–(5) are established. These allow the computation of hydrogeological parameters elsewhere within the study area based on the measured VESs. The equations are specific to the various geological settings observed in the region and the calculated values of T and K are also shown in Table 1.

Fig. 2
figure 2

Correlation between hydraulic conductivity and formation factor in a metamorphic rocks and b sedimentary rocks as well as between the transverse resistance and transmissivity in the same geological formations, respectively, c and d

Full size image

Results and discussion

Correlation between hydraulic parameters and formation factor led to linear relationships expressed in Eqs. (2)–(5).

  • Expression of the hydraulic conductivity as a function of the formation factor for metamorphic rocks:

    $$ K = - 0.0132*F + 0.8749 $$
    (2)

  • Expression of the hydraulic conductivity as a function of the formation factor for sedimentary rocks:

    $$ K = - 0.0695*F + 1.5499 $$
    (3)

  • Expression of the transmissivity as a function of the transverse resistance for metamorphic rocks:

    $$ T = 0.0631*TR + 3.0461 $$
    (4)

  • Expression of the transmissivity as a function of the transverse resistance for sedimentary rocks:

    $$ T = 0.0111*TR + 33.521 $$
    (5)

The transmissivity and hydraulic conductivity of the regional aquifer have been calculated using Eqs. (2)–(5) and are mapped as shown in Fig. 3, which also indicates the spatial distribution of other key data useful for groundwater exploration in the context of man-made wells. These parameters are aquifer resistivity, thickness and depth summarized in Table 2 with regard to geological formations.

Fig. 3
figure 3figure 3figure 3

Thematic maps of a transmissivity, b hydraulic conductivity, c resistivity, d thickness and e depth

Full size image
Table 2 Summary of the Adamawa aquifer characteristics derived from VES interpretation and pumping test
Full size table

Low values of the hydraulic conductivity are observed in the western part of the study area as well as in the eastern part, corresponding to granitic formations, where values are ranged between 0.09 and 0.19 m/day. On the other hand, higher values are observed within volcano-sedimentary formations. A comparison of the transmissivity and hydraulic conductivity maps (Fig. 3a, b, respectively) shows that those parameters are inversely proportional. The more an aquifer is transmissive the less it is hydraulically conductive. Resistivity values are ranged from 3 to 83 Ω m (Fig. 3c). Aquifer thickness ranges from 1.4 to 16.2 m (Fig. 3d). Aquifers located on the granitic formation in the western part of the region show lower thickness whereas those located in the eastern part as well as on sedimentary formations are thicker. The thickness is an important parameter to predict a sustainable and groundwater lasting exploitation. The aquifer depth is very important for two main reasons. In a context of man-made wells, this parameter helps population to select a site where the resource is easily accessible. In this situation, it is desirable to have very low depth values as in the middle part of the study area where those values are ranged from 10 to 28 m (Fig. 3e). On the other hand, shallow aquifer can be very vulnerable to various pollution due to infiltration. In this case, deeper aquifers are the most desirable ones.

Previous computation of transmissivity values did not consider the geological disparity. Yet, the region is huge and exhibits evident geological discrepancy. To address these previous shortcomings, the present paper (1) combines CMM and ISM to interpret VES, (2) performed further pumping tests and (3) take the geological discrepancy of the region into account. As a result, regression equations used to compute hydraulic conductivity as well as transmissivity from the electrical data differ from one geological formation to another, hence these results are more accurate. In fact, the geological setting is a very important factor in such a survey. A prediction equation becomes unreliable when exporting from a particular geologic environment—where it was established—to another.

The geological setting of this region is not uniform. Its variation affects hydro-parameters of the aquifer. Actually, "Pan-African" which was used to label the zone in the previous investigations refers to a regional stratigraphic and plate tectonic setting. Yet, aquifer in the tropics is rather connected to the weathering zone (ferralsols) and the paleoprofile (plinthites in former topographical lows) both of which are related and influence the thickness of the weathering zone which is related largely to the exhumation fracture system of the protolith. For this, “Adamawa aquifer” fits better than "Pan-African aquifer".

Contrarily to previous investigations where hydro-parameters were interpolated over the study area as a whole, the present paper takes into account the size and the geologic inhomogeneities of the area, and interpolate those parameters based on geologic subset. Statistical analysis was then performed per subsets. This approach improved and updated the hydro-parameters of the Adamawa aquifer.

In the context of sustainable development, VES can be considered as one of the most popular methods used for groundwater exploration. It is generally used for both groundwater quality and aquifer geometry. Various electrical and hydrogeological investigations were carried out in the study area with the purpose of constructing the subsurface lithology, major hydrogeological conditions, and the principal aquifer parameters (Aretouyap et al. 2015, 2017, 2018, 2019a, b; Bisso et al. 2019). Field data have been reinterpreted for determining aquifer resistivity and thickness using either the curve-matching method (CMM) or the inverse slope method (ISM). Values of hydro-parameters of the Adamawa aquifers, derived from various recent investigations, show slight differences. Those variations reflect differences between methods used in data collection and processing, as well as in geological conception. Other contributing factors include database size, quality, density and spatial distribution. In the present investigation, those differences will be investigated to give improved and updated results.

Correlations between electrical data and hydrogeological parameters are plotted in Fig. 3. The hydraulic conductivity is a decreasing linear regression of the formation factor for metamorphic and sedimentary contexts. On the other hand, the linear regression is increasing for the metamorphic formations. Those linear regressions are built with a minimum acceptable statistical number of points because of the non-geological uniformity of the study area. By establishing a linear regression for each geological formation, this reduced the number of experimental points to be used in each relationship. Further explanation for this variation difference may require additional investigation like rock compactivity, etc. It is necessary to conduct supplementary survey for this purpose.

Spatial distribution of hydrogeological parameters derived from electrical data are mapped in Fig. 3. In volcano-sedimentary formations, high transmissivity values are found in the NNE and the centre of the study area. This geological settlement has the highest transmissivity values in the overall study area. On the other hand, lowest transmissivity values are found in the metamorphic formations located the north and south parts of the study area. Similarly, high and intermediate values of the hydraulic conductivity are associated with both high and low transmissivity values, respectively, in volcano-sedimentary and metamorphic zones. Fundamentally, the transmissivity is the product of the aquifer thickness and hydraulic conductivity. Hence, high values of transmissivity match with either with thick aquifer or high hydraulic conductivity, contrarily to low transmissivity values. However, as for linear regressions, further investigation on the physical characterisation of local rocks would deepen the discussion of these results.

Conclusion

A series of hydro-geophysical investigations have been previously conducted in the Adamawa region to improve water supply to local population. The resulting parameters varied from an investigation to another because of the difference between methods and datasets used. This paper therefore addressed those dissimilarities by enhancing datasets and combining relevant methods. By augmenting the number of experimental VESs and combining CMM and ISM to interpret VES data taking into account the geological settings, this study led to updated values of aquifer transmissivity, hydraulic conductivity, resistivity, thickness and depth. In addition, “Adamawa aquifer” is more suitable to label this aquifer than “Pan-African” which refers to a regional stratigraphic and plate tectonic setting. For future geophysical investigation with the purpose of characterizing aquifers, it is necessary to subset the study area according to existing geological groups and to combine two or more interpretation methods.