Introduction

In the study area, surface water is a major source of water for domestic purposes, but due to challenges of population growth, climate change, and contamination from anthropogenic sources, its potentials have been pushed to its very limit (Opara et al., 2020; Urom et al., 2021). Groundwater is the second largest freshwater reservoir in the world, accounting for 12% of the world’s freshwater reserve, the largest resource being ice-locked water (87%), while surface water accounts for just around 1% of the world’s freshwater reserves (Gleick, 2011). Groundwater presents itself as a viable and safe source of potable water and a widely accepted and better alternative to surface water resources (McDonald et al., 2002; Singh, 2007). The search for groundwater in the study area was intensified because of the dearth of clean and potable surface water as most surface water across the study area are either saline, contaminated by mining activities, or infested with coliform and other pathogens (Obarezi & Nwosu, 2013; Obiora et al., 2015). Most surface water within the study area over the years have been plagued by Guinea worm which has further compounded the status of the surface water (Aghamelu et al., 2013; Okoronkwo, 2003). Also, the availability and productivity of groundwater in boreholes within the study area are usually problematic because most of the boreholes drilled are either abortive, unproductive, or have extremely low yields.

Successful exploration, exploitation, and effective management of groundwater resources therefore require an adept knowledge of the aquifer conditions including their geometrical and hydraulic parameters (Amos-Uhegbu, 2013; Ezeh, 2012; Hasan et al., 2020; Ogbuagu et al., 2018). These aquifer hydraulic parameters include transmissivity and hydraulic conductivity values. The conventional means of determining these parameters are usually through pumping test (Butler et al., 1999), but this approach is usually expensive and may be challenging in places where wells are widely spaced; thus, the interpolation of aquifer properties between the wells is usually difficult and often incorrect, since geological conditions vary relatively over very small distances (Bogoslovsky & Ogilvy, 1977; Muldoon & Bradbury, 2005). Vertical electrical sounding (VES) is an alternative means of estimating hydraulic properties of the groundwater system before drilling (Ekwe & Opara, 2012; Mbonu et al., 1991; Opara et al., 2020; Ugada et al., 2013). The integration of hydraulic parameters evaluated via pump testing in nearby monitoring wells and aquifer resistivity parameters estimated through geo-electrical techniques has been fully achieved by several authors (Chen et al., 2001; Dasargues, 1997; Ejiogu et al., 2019; Ekwe et al., 2020; Frohlich et al., 1996; Harry et al., 2018; Hasan et al., 2020; Heigold et al., 1979; Kalinski et al., 1993; Kelly & Frohlich, 1985; Mbonu et al., 1991; Nwosu et al., 2013; Ponzini et al., 1984; Purvance & Andricevic, 2000; Sinha et al., 2009; Ugada et al., 2013). Ugada et al. (2013) made use of the Dar Zarrouk parameters to estimate the aquifer properties of Umuahia. Ngwoke (2013) determined aquifer parameters in Ishiagu, Ebonyi State, using geo-electric methods. Also, Ekwe et al. (2020) determined aquifer parameters from geo-sounding data in parts of the Afikpo sub-basin, southeastern Nigeria. However, Sinha et al. (2009) proposed a hydrogeological model of the relationship between geo-electric and hydraulic parameters of an anisotropic aquifer.

Also, analytical equations generated by the integration of surface resistivity techniques and pumping test data had been used to estimate aquifer hydraulic parameters in different parts of Nigeria by some authors (Ejiogu et al., 2019; Emberga et al., 2021; Opara et al., 2020; Urom et al., 2021). These studies suggested that the estimation of hydraulic parameters from geologically constrained geo-electrical equations is feasible. However, such a relationship depends on specific areas and may have limited application in other areas except in areas of similar geology (Hasan et al., 2019; Purvance & Andricevic, 2000; Rehfeldt et al., 1992; Salem, 1999; Urom et al., 2021). An empirical equation that is formation-specific and constrained by the geology of the study area was proposed and used in the present study. The predictive accuracy of the model derived from the present study was increased by carrying out parametric soundings at locations with existing monitoring wells from which pumping test data were acquired. This was done to avoid overestimating or underestimating the predicted aquifer hydraulic parameter values (Opara et al., 2020).

Conventionally, the only direct method of estimating aquifer parameters is the pumping test technique. However, in most developing countries of the world, there is a serious dearth of pumping test data due to the huge cost of this very important analysis. To solve this problem, some classical publications have been made on how to estimate aquifer parameters from geophysical methods (e.g., Heigold et al., 1979; Niwas & Singhal, 1981, etc.). However, both the Heigold et al. (1979) and Niwas and Singhal (1981) equations generally used in the area to estimate hydraulic parameters from resistivity data were generated using data from overseas in areas with little or no relationship with the geology of the study area. The present study which is centered on alternative means of estimating aquifer hydraulic characteristics in areas with limited pumping test data using surficial resistivity methods therefore proposed and used a set of new empirical models together with the Heigold et al. (1979) and Niwas and Singhal (1981) equations. These new sets of models were generated with empirical data from the study area and are therefore constrained by the local geology of the area. The various model equations were therefore comparatively used and ranked to know the best alternative model equations that can be used to estimative aquifer hydraulic parameters from resistivity data on a regional scale when pumping test data are scarce or not readily available.

The idea behind this therefore is to improve the predictive capacities of the empirical equations used to estimate aquifer hydraulic characteristics from resistivity data. The objective is to provide an empirical relationship that is formation-specific, i.e., based on the local geology of the area because it is believed that incorporating the effect of local geology will improve the quality of the predictions using resistivity data. This study therefore aims to establish a relationship between aquifer parameters (hydraulic conductivity and transmissivity) and electrical resistivity-related parameters (aquifer resistivity, transverse resistance, etc.) and to make use of this relationship to estimate aquifer hydraulic parameters in areas with a paucity of pumping test data.

Location and geology of the study area

The study area which is in southeastern Nigeria lies between latitude 6˚ 4′ 76″ N and 6˚ 11′ 94″ N and longitude 7˚ 58′ 32″ E and 8˚ 9′ 99″ E (Fig. 1) and occupies an area of 442.57 km2. The fieldwork which involved field surficial electrical resistivity data acquisition took place between the 20th and 24th September 2019.

Fig. 1
figure 1

Accessibility and drainage map of the study area

Full size image

Based on the works of Reyment (1965), the study area falls within the Asu River Group formed during the Albian age and was folded into a northeast trend known as the Abakaliki Anticlinorium. Agumanu (1989) subdivided the Asu River Group based on stratigraphy into the Ebonyi Formation and Abakaliki Formation. The Ebonyi Formation (Mid-Albian) is underlain by the Abakaliki Formation (Late Albian–Cenomanian). The Ebonyi Formation dominates the eastern axis of the study area, which is made up of shales, rapid alternations of sandstones, siltstones, wacke stones, oolithic and serpulid stones, and mudstones (Fig. 2) (Oli et al., 2020).

Fig. 2
figure 2

Geologic map of the study area showing the VES and well locations

Full size image

The eastern axis of the study area on the other hand falls within the Abakaliki Formation, which is mostly dark-gray to black shales, and mudstones interspersed with siltstones, small feldspathic sandstones, and black micritic limestones. The stratigraphy of this formation indicates a reducing depository condition and anoxic environment, which aligns with Agumanu’s (1989) concept of formation. The sandstones occur as minimal litho-facies or lenses.

Methodology

Pump testing was carried out in a total of five (5) wells in the study area to determine the aquifer hydraulic parameters. The constant rate pumping method with a single well was adopted, with drawdown observations on the same well. The static water level was measured before the start of the pumping test using the electrical water level probe (dipper). A 1.5 Hp submersible pump was installed into the well, and pumping was done for 180 min. Dynamic water levels in the boreholes were measured at stopwatch intervals. After pumping was stopped, residual drawdowns were also measured at different time intervals.

Also, thirty-five (35) sounding points were selected in the study area with a parametric sounding performed at each of the wells where the pumping test was conducted, with the aid of an ABEM Terrameter (SAS 4000). The sounding points were geo-referenced using a handheld Global Positioning System (GPS). The VES data acquisition was executed using the Schlumberger array, with a maximum half-current (AB/2) electrode separation of 150 m and half-potential (MN/2) electrode separation of 15 m. Apparent resistivity (ρa) values were deduced from the observed field data using Eq. (1):

$${\rho }_{a}= \pi \left(\frac{{a}^{2}}{b}-\frac{b}{4}\right)\frac{\Delta V}{I}$$
(1)

Estimation of geo-hydraulic parameters

Estimates of geo-hydraulic parameters from pumping test

The Cooper and Jacob solution method was used to determine the aquifer-derived parameters (transmissivity and hydraulic conductivity) from the pumping test. This was achieved using a computer software (Aquifer Win32) by plotting drawdown against their respective time data acquired in the semi-log format during the pumping test. The transmissivity values were calculated using the formula by Freeze and Cherry (1979) as shown in Eq. (2):

$$T= \frac{2.3Q}{4\pi \Delta S}$$
(2)

where T = transmissivity in m2/day, Q = discharge rate in m3/day, and ΔS = change in drawdown over one logarithmic cycle.

The hydraulic conductivity was calculated from the transmissivity and aquifer depth values, which is, in this case, assumed to be the length of the screen, using the equation by Freeze and Cherry (1979) as shown in Eq. (3):

$$K= \frac{T}{B}$$
(3)

where K = hydraulic conductivity in m/day, b = aquifer thickness in m, and T = transmissivity in m2/day.

Estimates from surficial resistivity data

Several electrical resistivity-based empirical equations have been previously used to estimate aquifer hydraulic and transmissivity values across the study area. These empirical equations include the equations of Niwas and Singhal (1981) and Heigold et al. (1979) and the proposed new model.

The determination of aquifer hydraulic characteristics can be accomplished by using parameters of transverse resistance and longitudinal conductance from Dar-Zarrock parameters. Niwas and Singhal (1981) developed, on one hand, an empirical relation between transmissivity and transverse resistance and, on the other, longitudinal conductance and transmissivity. Based on Darcy’s law, the fluid discharge Q is given by Eqs. (4) and (5):

$$Q=KIA$$
(4)

And from Ohm’s law

$$J= \delta E$$
(5)

where K = hydraulic conductivity, I = hydraulic gradient, A = cross-sectional area perpendicular to the direction of flow, J = current density, E = electric field intensity, and δ = electrical conductivity (inverse of resistivity).

Considering a prism of an aquifer material having a unit cross-sectional area and thickness h, Niwas and Singhal (1981) combined Eqs. (4) and (5) to get the equation given in Eq. (6):

$$T=k\delta R=KL/\delta$$
(6)

where T = aquifer transmissivity, R = transverse resistance, \(\updelta\) = aquifer conductivity, and L = longitudinal conductance.

It is well documented that quantitative representations of vertical electrical sounding data contribute to the creation of geo-electric layers in resistivity measurements. Layer parameters like aquifer depth and thickness therefore can be better identified with information from geo-electric layers. The resulting layer parameters are usually used to determine the Dar-Zarrock parameters. Therefore, the product of the aquifer’s apparent resistivity (ρ) and the aquifer’s thickness (h) results in transverse resistance (R) as shown in Eqs. (7) and (8):

$$R=hp$$
(7)
$${K}_{NS}= k\delta \rho$$
(8)

Niwas and Singhal (1981) maintained that areas with similar geologic settings and water quality usually have fairly constant diagnostic constants (diagnostic constants is the product of the hydraulic conductivity (k) from pumping test and the electrical conductivity (\(\updelta\)). Based on this, therefore, the aquifer hydraulic parameters which vary spatially across an area both for the areas with pumping test values and areas without wells can be estimated from resistivity data measured at the surface of the earth.

Also, the Heigold et al. (1979) equation was used in this study to estimate hydraulic parameters across the study area. The Heigold et al. (1979) empirical equation is based on the relationship between hydraulic conductivity (K) obtained from pumping test from monitoring wells and water resistivity estimated from resistivity data carried out close to the wells as shown in Eq. (9):

$${K}_{HG }={386.40Rw}^{-0.93283}$$
(9)

where Rw is aquifer resistivity. Then, the transmissivity of the aquifer (T) can now be estimated using the relationship given by Niwas and Singhal (1981) in Eq. (10):

$$T=k\delta T= ks/\delta =kh$$
(10)

where δ is the electrical conductivity (inverse of resistivity) and S is the longitudinal conductance.

Finally, a new set of formation-specific empirical equations that has a relationship with the intrinsic rock properties in the study area were proposed and used in the present study. Using the empirical relationship established between hydraulic conductivity derived from the pumping test in the study area and aquifer resistivity on one hand and that between transmissivity and transverse resistance, a set of two formation-specific model equations that are geologically constrained and sensitive were generated. Hydraulic conductivity and transmissivity acquired from the wells where pumping tests were conducted were plotted against aquifer resistivity and transverse resistance values, respectively, obtained from parametric soundings at the well locations in the different formations (Fig. 3a, b, c, and d), which thereafter were used to estimate transmissivity and hydraulic conductivity at locations where pumping test was not conducted.

Fig. 3
figure 3

Cross-plots showing relationships between aquifer hydraulic parameters and VES estimated parameters: a Kebfm, b Kafm, c Tebfm, d Tafm

Full size image

These cross-plots yielded two sets of novel empirical equations of hydraulic conductivities (K) and transmissivities for Ebonyi and Abakaliki Formations, respectively, as given in Eqs. (11)–(14):

$${{K}^{ebfm}=4.1559Rw}^{-0.319}$$
(11)
$${{K}^{afm}=0.0114Rw}^{0.7792}$$
(12)
$${T}^{ebfm}={5330.4R}^{-0.928}$$
(13)
$${T}^{afm}={0.0092R}^{0.8117}$$
(14)

where Kebfm = hydraulic conductivity for the Ebonyi Formation, Kafm = hydraulic conductivity for the Abakaliki Formation, Tebfm = transmissivity for the Ebonyi Formation, Tafm = transmissivity for the Ebonyi Formation, Rw = aquifer resistivity, and R = transverse resistance. The coefficient of determination (R2) for Kebfm, Kafm, Tebfm, and Tafm was found to be 1.0, 0.997, 1.0, and 1.0, respectively, exhibiting a very strong positive relationship between the parameters.

Results and discussion

Interpretation of layer parameters

VES data were used to extract interpreted curves (Fig. 4). Interpretation of the geo-electric curves across the study area revealed four to seven (4–7) geo-electric layers with different intra-facies and inter-facies changes (Table 1). The curve types were observed to be mainly of the QH, QHK, QHKH, QQH, KHK, QHAK, and QQHK types. Ngwoke (2013) stated that the existence of several curve types shows a non-uniformity of resistivity patterns across the study area. The non-uniformity of layering and modification of layer properties is due to differential weathering, fracture anisotropy, and other geological factors, which generally result in differences in resistivity trends across the area of study. The dominant curve type is the QH curve with approximately 37%, QHK with 23%, and HK type with 9%, with the QQH, KHK, and QHAK accounting for 6%, respectively, while QQHK, KH, HA, and QHK each account for 5%.

Fig. 4
figure 4

Sounding curves from a VES 4, b VES 7, c VES 18, d VES 20

Full size image
Table 1 Summary of interpreted layer parameters from the study area
Full size table

Aquifer hydraulic parameters

The results of aquifer hydraulic parameters acquired using the pump testing techniques in the five wells are presented in Table 2. The pumping test data were analyzed and plotted using Copper–Jacob straight line curve with the aid of Aquiwin-32 software. Sample plots of the processed pumping test data acquired from the study area are presented in Fig. 5.

Table 2 Computed hydraulic conductivity and transmissivity values derived from different models
Full size table
Fig. 5
figure 5

Pumping test curves analyzed using Cooper and Jacob method for a Ekka Ezza, b Onueke market, c Ndiechi Ndufu Achara, d Ishieke Ndufu Igbudu

Full size image

Aquifer hydraulic conductivity (K) estimates of the study area

Hydraulic conductivity (K), which is a measure of the ease with which a fluid will pass through a medium, and transmissivity (T), which is the rate of flow of fluid under a unit hydraulic gradient through a unit width of the aquifer of thickness, were estimated using the Niwas and Singhal (1981) (KNS) equation, Heigold et al. (1979) (KHG) equation, and the new empirical equations as shown in Table 3.

Table 3 A paired t test for the different models
Full size table

Hydraulic conductivity values estimated from the Heigold model using (Eq. (11)) for the Ebonyi Formation vary from 0.75 to 22.6 m/day, with a mean value of 5.84 m/day, while that of the Abakaliki Formation varies from 1.78 to 39 m/day and has a mean value of 17.8 m/day. From the hydraulic conductivity map (Fig. 6a), the areas underlain by the Abakaliki Formation (Eastern axis) have a higher value compared to those areas underlain by the Ebonyi Formation (western axis). This is in agreement with the geology of the study area as previously explained by Agumanu (1989). Generally, across the study area, shales dominate the Abakaliki Formation and usually have a lower hydraulic conductivity when compared with the Ebonyi Formation, which has an alternating sequence of sandstones, siltstones, and shales. Using the Niwas and Singhal (1981) empirical equations, aquifer hydraulic conductivity was estimated by taking the product of the diagnostic constant (\(\mathrm{k\delta })\) and aquifer resistivity (\(\uprho\)) at VES locations as shown in Eq. (8). The average diagnostic constant of 0.00721 was used for areas underlain by the Ebonyi Formation, while areas underlain by the Abakaliki Formation have a mean diagnostic parameter of 0.00352. The estimated hydraulic conductivity of the study area for the Ebonyi Formation ranges from 0.15 to 5.87 m/day, with a mean value of 1.32 m/day. For the Abakaliki Formation, which is overlain by the Ebonyi Formation, the estimated hydraulic conductivity ranges from 0.04 to 0.61 m/day, with an average of 0.25 m/day. Areas with higher aquifer hydraulic conductivity usually have higher hydraulic connectivity and permeability and are generally associated with higher groundwater potential (Opara et al., 2020). The hydraulic conductivity map generated from the estimates predicted using the Niwas and Singhal model is shown in Fig. 6b.

Fig. 6
figure 6

Contour map of the study area showing hydraulic conductivity, m/day: a Heigold model, b Niwas and Singhal model, c model derived from the present study

Full size image

Also, Eqs. (11)–(12) which represent the new model equations proposed and used in this work were used to estimate the hydraulic conductivity values of the Ebonyi and Abakaliki Formations within the study area. Hydraulic conductivity values estimated using the new model for areas underlain by the Ebonyi Formation range from 0.49 to 1.5735 m/day with a mean value of 0.9205 m/day, while those underlain by the Abakaliki Formation have hydraulic conductivity values that vary from 0.0775 to 1.3023 m/day, with a mean value of 0.2883 m/day. There is a high level of agreement between the hydraulic conductivity estimated from the pumping test and that from the new model derived from the present study when compared with Niwas and Singhal and Heigold model as shown in Table 2. This shows that the model equation proposed and used in the present study which is geologically constrained is more effective in estimating aquifer hydraulic parameters across the study area. From the hydraulic conductivity contour map of the study area generated from values estimated using the new model (Fig. 6c), there exists a hydrogeological divide with the Ebonyi Formation in the western axis of the study area having higher hydraulic conductivity values and therefore a more prolific aquifer system than the Abakaliki Formation which is in the eastern axis of the study area with lower hydraulic conductivity values. These findings are in agreement with previous works done in the study area (Agumanu, 1989; Ekwe et al., 2015; Oli et al., 2020). Within the Abakaliki Formation, areas with hydraulic conductivity greater than the surrounding formation are believed to be associated with highly fractured shale zones which improved the porosity and permeability of the formation.

Estimation of aquifer transmissivity (T) of the study area

Aquifer transmissivity estimated across the study area using the new model ranges between 0.29 and 57.27 m2/day with a mean value of 6.59 m2/day. The transmissivity values within the area underlain by the Ebonyi Formation vary from 0.63 to 57.27 m2/day with a mean value of 8.23m2/day, while that of the Abakaliki Formation ranges from 0.29 to 9.22 m2/day with a mean value of 3.44 m2/day. The contour map of the transmissivity values estimated using the new model is shown in Fig. 7a. Also, Niwas and Singhal’s model was also used to estimate transmissivity across the study area as shown in Eq. (10) by using the product of the aquifer hydraulic conductivity estimates made from the Niwas and Singhal (1981) equation and the aquifer thickness. The estimated values for the Ebonyi Formation therefore ranges between 0.95 and 124 m2/day with a mean value of 20.19 m2/day, while that of the Abakaliki Formation ranges from 0.25 to 17.5 m2/day with an average of 5.54 m2/day. Based on these predictions, therefore, the Ebonyi Formation has higher transmissivity values than the Abakaliki Formation as shown in Fig. 7b. Finally, the aquifer transmissivity values estimated by multiplying the hydraulic conductivity values estimated using the Heigold model by the thicknesses of the aquifer for the Ebonyi Formation range from 3.01 to 934 m2/day with a mean value of 142 m2/day, while that of the Abakaliki Formation range from 50.2 to 1347 m2/day with a mean value of 507 m2/day, with the map shown in Fig. 7c. Analysis of the transmissivity contour map of the study area, estimated by using the Heigold model (Fig. 7c), suggests that areas underlain by the Ebonyi Formation have a lower transmissivity than areas underlain by the Abakaliki Formation. This in particular is not in agreement with the geology of the area, thereby showing that the Heigold model is defective for the study area. Heigold et al. (1979) equation therefore typically under-predicts areas which are not similar geologically to the study area from where the empirical equation was generated.

Fig. 7
figure 7

Contour map of the study area showing transmissivity in m2/day: a model derived from the present study, b Niwas and Singhal model, c Heigold model

Full size image

Statistical analysis was carried out to ascertain the reliability of the different empirical equations/models in estimating hydraulic conductivity by comparing them with the values from the widely accepted pumping test technique. A paired t test was used to compare the values of the standard deviation, mean, variance, and Pearson correlation of the various hydraulic conductivities estimated from other models with those from the pumping test as shown in Table 3. From Table 3, it was observed that K values estimated from the new model equations when compared with K values from the pumping test revealed a Pearson correlation of 99%. This represents a strong positive correlation. The other models (KNS and KHG) presented a strong negative correlation with that from the pumping test. The observed mean difference of hydraulic conductivity estimated from Niwas and Singhal (1981) equation, Heigold et al. (1979) equation, and the new model equation when compared with the values of the pumping test showed that the new model values have a lower observed mean difference than the others (Table 3). This validates the efficiency of the model derived from the present study in estimating hydraulic conductivity when there is dearth of pumping test data.

Groundwater potential

The groundwater potential of the study area was assessed based on the transmissivity of the aquifer at each sounding point estimated using the new model. Krasny’s (1993) classification of transmissivity magnitude as shown in Table 4 was used to assign groundwater supply potentials of the various locations in the study area. Based on Table 5, it was observed that the aquifer potentials of the study area range from low to intermediate. The groundwater potentials at two (2) of the locations representing 6% of the study area have groundwater potential which can only sustain limited consumption, with twenty-nine (29) of the locations which represent 83% of the study area capable of providing groundwater potentials that can serve for private consumption, while the remaining four (4) locations which represent 11% of the study area hold a groundwater potential that can serve as a local water supply. These areas that can sustain local water supply are dominated by areas underlain by the Ebonyi Formation. The aquifer potential map of the study area is shown in Fig. 8.

Table 4 Classification of transmissivity magnitude (After Krasny, 1993)
Full size table
Table 5 Transmissivity classification based on data collected in the study area
Full size table
Fig. 8
figure 8

Groundwater potential map of the study area

Full size image

The results of this study have helped to delineate the groundwater potential zones within the study area. Evidently, the findings of the present study thus revealed a groundwater divide in line with the geology of the study area with the Ebonyi Formation having a higher groundwater potential than the Abakaliki Formation. The findings of the present study are in agreement with the results of previous studies within the study area (Ekwe et al., 2020; Obiora et al., 2015; Oli et al., 2020).

Conclusion

The present study has clearly demonstrated the effectiveness of the application of surficial resistivity data in aquifer hydraulic estimation. Aquifer hydraulic parameters including aquifer hydraulic conductivity and transmissivity were estimated using multiple resistivity-based empirical equations even in areas with a paucity of pumping test data. These analytical and empirical equations which have been used with fairly high level of success were improved by adopting formation-specific equations which were constrained geologically. Statistical analysis of aquifer hydraulic parameters estimated from the different models revealed that the new model proposed and used in the present study clearly showed values that have the closest relationship with values obtained from the pumping test. Transmissivity estimated from the new model suggested that areas underlain by the Ebonyi Formation have a greater aquifer potential when compared with those areas underlain by the Abakaliki Formation. This can be explained by the geology, as areas within the Abakaliki Formation with higher aquifer potential are suspected to be highly fractured shales. This is also validated by Krasny’s groundwater potential classification of the study area, with areas underlain by the Ebonyi Formation having greater groundwater prospects than those underlain by the Abakaliki Formation. Therefore, exploitation should be focused more on areas underlain by the Ebonyi Formation for a greater yield. The study therefore clearly revealed a pronounced groundwater divide between the Ebonyi and Abakaliki Formations of the study area.

The closeness of the estimated results obtained from the interpretation of the vertical electrical sounding results with those obtained from pumping tests from available borehole locations has further shown the validity of the present study. Electrical resistivity method is therefore a useful tool for understanding the aquifer systems in the study area. The study has shown that direct current electrical resistivity methods are not only useful in groundwater exploration or delineation of aquifer geometry but can also be effective in the estimation of aquifer hydraulic parameters.