Introduction

Water is a fundamental natural resource that exists as both surface water and groundwater. The development and growth of any society is contingent on the availability and utilization of appropriate water. Water covers almost two-thirds of the world's total land area (Agbasi and Etuk 2016). This essential resource, which is often sparse, plentiful, yet unevenly distributed in geography and time is critical to human survival. Groundwater is water that has penetrated into the soil directly via precipitation, recharge from streams and other natural water sources and man-made recharge (Ibuot et al. 2022). More than 1.5 billion people worldwide are reported to rely on groundwater for their drinking water needs. Groundwater occurrence, distribution, and flow are discontinuous and governed by the dynamic interplay of numerous environmental elements such as geology, slope, overburden thickness, and weathering (Agbasi et al. 2019; George et al. 2021).

Despite the fact that the nineteenth and twentieth centuries saw significant industrial and economic revolutions, many developing nations (including Nigeria) are still unable to provide enough potable water for their citizens (Oli et al. 2021). For example, with an expanding population and industrialization, demand for water resources in Nigeria is continually increasing, forcing residents to rely on drinking water from wells. More than 70% of accessible drinkable water in Nigeria is tapped from the ground, with the exception of Abuja, where more than 80% of the municipal area is connected to a pipe-borne water network (George et al. 2015a; b; Abdullateef et al. 2021; Ejepu et al. 2022). As a result, the importance of groundwater in supplying a bigger share of water demand cannot be overstated. Although it is possible to identify groundwater potential through observation wells, it is uneconomical or expensive in extensive areas. Many multivariate statistically based techniques, which do not require field work, have been used by researchers to estimate groundwater potential. Such techniques include decision tree model (Lee and Lee 2015), logistic regression model (Pourtaghi and Pourghasmi 2014) and the principal component analysis (PCA) (Helena et al. 2000). Alternatively, the diagnostic hierarchy process (AHP) is viewed to be a simple, consistent, valuable, obvious and dependable technique (Machiwal et al. 2011). The AHP approach is a straightforward and a matching technique that can be integrated with geographic information system (GIS) and remote sensing (RS) data for local and regional groundwater potential studies. Groundwater parameters such as precipitation, aquifers, land use, and soil type can easily be defined as spatial data in a GIS environment. Groundwater systems are made up of a complex web of interactions between the physical environment (soils, geology, precipitation, vegetation, and terrain) and the social and economic activities that put pressure on the physical environment (Ejiogu et al. 2019). The study is provoked by inadequate understanding about the geographical units responsible for groundwater occurrence, migration, and storage. As a result, geospatial groundwater assessment associated with qualitative and quantitative considerations is required to offer a better knowledge of groundwater dynamics for sustainable usage.

Several land and water management strategies have emerged, and remote sensing and Geographic Information Systems (GIS) have gained popularity. The availability of free-to-low-cost satellite imagery and digital elevation models is the most significant contribution of remote sensing technology to groundwater exploration (Abdulrazzaq et al. 2020; Ige et al. 2020). GIS is an important tool that can manage vast amounts of data (both geographical and spatial) and can enable a variety of quantitative and qualitative spatial analyses and interpretations (Aladeboyeje et al. 2021). This research project is primarily aimed at producing and characterizing groundwater potential maps that can classify the study area into separate zones based on their groundwater potential. To this end, the novelty in this work involves the use of local information from surface geophysical technique and regional/spatiotemporal information from GIS-based multicriteria decision making (MCDM) and AHP methods to produce groundwater-potential maps of the study area using logical groups of characterizing hydrogeological entities. The resulting detailed map of the area can be utilized as inputs for understanding the hydrogeological systems of the study area and as data for future research (Adiat et al. 2012; Ejepu et al. 2020; Aretouyap et al. 2022). The generated, collected, and digitized data were organized into logical groups of entities with characterizing geological factors (lithology and lineament), drainage density, and physiographic factors (slope, land use/land cover) according to Ekanem et al. (2022a) to produce a properly organized geo-database capable of allowing responsible officers, stakeholders and subsequent researchers to make decisions and reviews for water resource extraction, monitoring and management.

Location and geology of the study area

The digitized map of Fig. 1 shows that the study area covers Ibeno, Mkpat Enin, Onna and Eket Local Government Areas (LGAs) of Akwa Ibom State, which stretch out from longitudes 7.68° to 8.35 °E and latitudes 4.25° to 4.90 °N in the southern parts of Akwa Ibom state and are active shoreline regions within the Gulf of Guinea. Geologically, the study area is part of the Benin Formation, which consists of Coastal Plain sands. The general geology of the study area is primarily Coastal Plain sand (Benin Formation) deposits, and the landscape is averagely flat. It is a deposition of deltaic, estuarine, lagoonal, fluvial-lacustrine and marine materials by its locations (Short and Stauble 1967; George et al. 2014; Ibuot et al. 2013; George 2021a, b). A few LGAs have fairly undulating landscape occasioned by isostatic compensation provoked by gravitational pull energized by the Niger Delta Basin that is actively collapsing. These areas are as high as 60,000 mm above mean sea level owing to the impact of the Atlantic Ocean and its concomitant rivers and valleys (Ibanga and George 2016). The aquifer is made up of both closed and open hydrogeologic unit, with varied levels of salinity in groundwater. The continental flood plain sands and its alluvial deposits have modest to high rate of permeability and porosity. Salt water incursion into sedimentary aquifers occurs under the fresh water lens, which stretches out by about 5 km to the shore. The main hydrogeologic units in the region are Coastal Plain Sands interspersed with poorly sorted continental (fine-medium-coarse) sands and gravels. The suite of sands alternate with thinly clayey sands, lignite streaks and lenses (George 2021b). The spread of aquifer sands is cut off by the thin argillaceous layers. This gives rise to the formation of multi-aquifer systems throughout the region.

Fig. 1
figure 1

Map of the world, Nigeria, and the study location showing study locations with BH and VES points

Full size image

Sediments in the study area have grain sizes ranging from coarse to fine, and they are unevenly and poorly sorted, with a large thickness and a negligible quantity of clay, silt, and sandstone with various sequences that correspond to separate aquifers of varying thicknesses. The upper sandy aquifer has composition of alluvial materials such as gravel, fine- to medium-grained carbonaceous sand or unconsolidated sands of various grades (Reijers et al. 1997; Akpan et al. 2013).

Materials and methods

VES technique

The basic lithological sand-bed characteristics known as primary geo-electric indices in the study area were estimated using data gathered from vertical electrical sounding survey at different locations. The Schlumberger electrode configuration arrangement employed through the deployment of terrameter and its accessories to measure and record the subsurface resistances (George 2020; Aizebeokhaia et al. 2021) was chosen due to its ability sensitivity to the vertical and lateral section of the soil profiles. The study uses a surface geophysical survey from nineteen locations within the Akwa Ibom State coastal zone, which serve as potential prolific hydrogeological units for groundwater abstraction. The Schlumberger electrical resistivity configuration employed deduced the bulk resistivity through the direct measurement of resistance from the instrument by current injected into the ground (Inim et al. 2020; Thomas et al. 2020; George et al. 2022). The current potential fields (delivered to the ground) were measured using two pairs of electrodes, each corresponding to a potential electrode (MN) and current electrode (AB). For assurance of quality in the field measurements, straight lines were maintained in both the outer (current) and inner (potential) electrode separations. Besides, about one-third of the electrodes were driven into the earth with assured grip between the metal electrodes and the ground (Obinawu et al. 2011; George et al. 2016). MN varied from 0.5 m at AB = 2 m up to 40 m at AB = 600 m (Akpan et al. 2013; Ibuot et al. 2013).

Apparent resistance (Ra) measured from the Schlumberger electrode configuration arrangement was used with geometric factor \(G = \pi \left[ {\frac{{\left( {AB/2} \right)^{2} - \left( {MN/2} \right)^{2} }}{MN}} \right]\) to compute apparent resistivity ( \(\rho_{a}\)) values through Eq. (1):

$$\rho_{a} = \pi \left[ {\frac{{\left( {AB/2} \right)^{2} - \left( {MN/2} \right)^{2} }}{MN}} \right]R_{a}.$$
(1)

The processing of the measured resistivity data began by smoothening using the log–log plot of \(\rho_{a}\) against (AB/2), followed by electronic processing, which involved the use of WinResist software program (Vander Velpen and Sporry 1993). The data processed manually to remove outliers were electronically processed quantitatively using a 1-D least square computer-aided forward modelling software program known as WINRESIST. With the constraints of nearby ground truth data, the software program furnished the interpreted curves by distinguishing the primary geo-electric indices such as layer resistivity, thickness and depth, in which their respective aquifer values are shown in (Table 1). The goodness of fit established through the matching of field data with theoretical values iteratively was manifested in the achieved root mean square errors (RMSE) generally found to be < 5 (see Fig. 2).

Table 1 Summary of geo-sounding statistics in the study area
Full size table
Fig. 2
figure 2

Sampled correlations of VES 1, 2, 11 and 19 curves with their contiguous lithological log in the study area

Full size image

Aquifer, a body of rocks/or sediments that holds water, is drenched rocks or geological units through which volumes of fluid/water transmits into springs and wells. In nature, these trenched rocks possess high degree of variability, which practically describes its hydraulic properties (Obiora et al. 2016). The measurement of aquifer water levels is achieved through indirect measurement, which is used in determining the direction of ground water flow, and in measuring the aquifer hydraulic properties (George 2021a, b). Geo-electrical technology, which is cost-effective, compared to the traditional pumping test is intelligibly pragmatic with promising results (Ekanem et al. 2022a, b; Ikpe et al. 2022; Abdulrazzaq et al. 2022). These parameters are central in determining the flow of water in nature through the aquifer systems and their reaction to fluid extraction. The flow of groundwater through the strata and their release from storage within the strata govern the response to changes in their supply from the natural stockpile (Ekanem et al. 2020). The measurements of these parameters were achieved with the aid of aquifer samples.

Measurements from cored samples

Porosity

Cored samples collected during borehole drilling at sundry coastal points close to the VES points were employed to deduce porosity and hydraulic conductivity. The samples were sufficiently soaked in distilled water to remove any potential salt contaminant that could influence the pores of the geological samples and possibly the porosity determination. The clean samples were dried with oven so that extraneous fluid within the pores can be removed. Thereafter, the bulk density \(\left( {{\rm B}_{d} } \right)\) of each of the samples was estimated through the soil weighted mass \(\left( {Wm} \right)\)-volume equivalent \(\left( {VE} \right)\) ratio of the dry samples as given in Eq. (2):

$${\rm B}_{d} = \frac{Wm}{{VE}}.$$
(2)

With the soil particle density \(\left( {P_{d} } \right)\), ranging from 2.60 to 2.75 g/cm3 with constant mean value of 2.65 g/cm3 (Umoh et al. 2022), the fraction of aquifer pore space, otherwise known as porosity \(\left( \varphi \right)\), was estimated and is recorded in Table 2 for each location in the nearby VES station using Eq. (3):

$$\varphi = \left( {1 - \frac{{B_{d} }}{{P_{d} }}} \right).$$
(3)
Table 2 Measurement scale of AHP (Saaty 1980)
Full size table

Hydraulic conductivity

Using the same cored samples, the hydraulic conductivity, an important property of rock/soil units, was estimated using the constant flow head permeameter (Fig. 3) according to Todd (1980). Constant and continuous supply of water via the inlet provoked steady flow of water through sample with area A (Fig. 3). Stop clock was deployed to record the time (t) of water flow until the volume (V) flowing through the soil sample is 100 cm3 (Fig. 3). Here, different samples are believed to have varying rate of flow based on their pore space/porosity and the consequent permeability (Todd 1980; Obiora et al. 2015). The constant flow head permeameter allowed the determination of the hydraulic conductivities of unconsolidated different soils under the low heads. Figure 3 is made up of aquifer sample of length L and area of cross section A, sandwiched between two leaky plates in a cylindrical tube under a constant-head differential h. The steady water from reservoir supply entered the medium cylinder from the base and was collected after passing upward through the aquifer sample as overflow (Tang et al. 2017). Based on the setup, hydraulic conductivity Kh, which is a measure of the ease of a medium to transmit fluid (Agbasi and Etuk 2016; George 2020), was estimated for each aquifer sample. The values were computed according to Eq. (4) in m/s and converted to m/day as given in Table 2.

$$K_{h} = \left( {\frac{V \times L}{{A \times h \times t}}} \right).$$
(4)
Fig. 3
figure 3

Permeameter for measuring hydraulic conductivity of aquifer sample by constant flow head method (Todd 1980)

Full size image

Transmissivity (T)

According to Massoud et al. (2010), \(K_{h}\) and T values play a significant role in aquifer system categorization. These properties generally rate the ease of an aquifer to transmit water over a unit thickness of hydraulic conductivity and across the entire drenched thickness of the aquifer. Often times, they are estimated from pumping tests. Practically, saturated thickness and hydraulic conductivity are used to estimate transmissivity. The mean value of hydraulic conductivity of each layer was multiplied by the inverted thickness of aquifer from 1-D resistivity inversion according to Younger (2007) in order to obtain transmissivity recorded in Table 3.

$$T = K_{h} \times h$$
(5)
Table 3 Summary probable and economic aquifer system at various VES stations surveyed
Full size table

Remote sensing (RS) and geographic information systems (GIS)

The Satellite Shuttle Radar Topography Mission (SRTM) and Digital Elevation Model (DEM) data were used for the spatiotemporal investigation. The photographs and datasets were chosen depending on their availability. The satellite gathers photographs of the Earth on a 16-days repetition cycle, using the global reference system-2 as a reference (Aladeboyeje et al. 2021). The scene spreads approximately 170 km north–south by 183 km east–west. Landsat 7's ETM + sensor improved on previous Landsat instruments by adding two new spectral bands: a deep blue visible channel (band 1) specifically designed for water resources and coastal zone investigation; and a new infrared channel (band 9) for detecting cirrus clouds. Two thermal bands (TIRS) collected data with a minimum resolution of 100 m, although they are registered and transmitted with the 30-m data package. ArcGIS 10.8 was used to produce and combines several thematic layers required for identifying relevant groundwater potential locations as well as conducting groundwater studies.

Preprocessing of GIS/RS

The remote sensing data were geometrically adjusted prior to analysis. The bands and pictures from Landsat 8 were layered and mosaicked. The mosaicked picture was clipped using the extract tool in ArcGIS through the Akwa Ibom state shapefile. As with Landsat 8, the SRTM DEM was produced as a mosaic and subset of the research region.

Preparation of the thematic layers

The number of thematic layers employed is determined by the availability of data in the research region. Five thematic layers, including slope, land use/cover, soil type, drainage and lineament density, were identified as efficient parameters to estimate groundwater potential zones.

Slope map

A slope is a geomorphologic characteristic that influences the infiltration and recharging of a groundwater system in a specific location, and it is measured in degrees. In terms of groundwater, flat locations with low slope store rainfall, which enables recharging, whereas elevated places with high slope have considerable run-off and poor infiltration (Makonyo and Msabi 2021), leading to low groundwater. The slope map for the research region was created using the spatial analyst tool in the ArcGIS 10.8 environment, and it was afterwards classified prior to the overlaying procedure for groundwater mapping.

Land use/cover

Another factor influencing groundwater occurrence is land use and cover. Land cover/use has an influence by either lowering runoff or facilitating or by retaining water on their leaves. Water droplets caught in this manner fall to the earth and refresh the groundwater. Land use/cover may also have a detrimental impact on groundwater through evapotranspiration, assuming steady interception (Biswajit et al. 2020). The land use and coverage of the study have a combination of built-up, vegetation, barren land, agricultural area, and water bodies. Visual interpretation was used to create an overview of the land use and land cover. For the final land-use map, supervised classification (maximum likelihood classifier) was used using the proper band combinations. The resultant map was then classed in order to delineate groundwater potential.

Soil type

The lithology, stratigraphy, and structure of the geological strata present affect the presence of groundwater as well as the extent and distribution of aquifers and aquitards in an area. The porosity and permeability of aquifer rocks are both influenced by soil type. The soil type map for the research region was digitized from an Akwa Ibom state geology map and classed in preparation for groundwater mapping.

Drainage density

Drainage, which is plainly apparent on remote sensing imaging, is one of the most significant indicators of groundwater potential because the underlying lithology controls drainage pattern, texture, and density in a fundamental way (Falebita et al. 2020; Omosuyi et al. 2021). The drainage pattern reflects the pace at which precipitation infiltrates in comparison with surface runoff. High drainage density suggests an adverse location for groundwater existence, moderate drainage density indicates intermediate groundwater potential; and less or no drainage density indicates a good site for groundwater existence. The drainage pattern can be created using the SRTM DEM, and a drainage density map can be realized. The map generated can then be categorized to delineate groundwater. The drainage density \(\eta\) per kilometer used in geospatially estimating the map in the study is given by Eq. (6):

$$\eta = \frac{L}{A},$$
(6)

where L is the total length of channel in kilometer and A is the areal expanse of the study in square kilometer.

Lineaments

Lineaments and their intersections provided information on the occurrence, transport, and storage of groundwater and are hence useful guides for delineation of groundwater potential. Many groundwater exploration operations in many nations have recently had greater success rates when drilling locations were guided by lineament mapping (Arshad et al. 2020). Lineament density maps were created and classed using the spatial analyzer tool.

Weighting of thematic maps using the Analytic Hierarchy Process (AHP)

AHP is a basic mathematical matrix-based approach that allows for intuitive determination of the relative weight of numerous criteria (Mogaji et al. 2021). AHP streamlines the decision-making process by letting the decision-maker focus on one aspect of the choice at a time (Ifediegwu 2022; Ghosh et al. 2022). This simplicity streamlines the decision-making process and makes it more transparent. AHP therefore provides a method of organizing decisions (Table 2), priorities, and judgments by allowing the decision maker to arrange them in a hierarchy with the highest priority, or objective, at the top. It also includes systematic checks on judgment consistency, which is one of its main advantages over other multi-attribute value systems (Ejepu et al. 2017; Gohsh et al. 2022; Thapa et al. 2017). This consistency testing allows for the comprehension of inconsistent judgments. Because the method is adaptable, the decision maker can modify their priorities to ensure consistency. A matrix with a consistency ratio (CR) that is greater than 0.1 is often required for re-evaluation. All discovered themes in this study were evaluated against each other in a pairwise comparison matrix, which is a metric to represent the relative preference amongst the parameters in terms of groundwater occurrence, transportation, and storage as given in Table 2.

Normalization of assigned weights utilizing AHP (Wind and Saaty 1980) on Saaty’s scale examined two themes and classes at a time based on their respective relevance to identify groundwater potentials and recharging zone. Following that, AHP was used to build pairwise comparison matrices with specified weights to various thematic layers and classes, and the weights were normalized using the eigenvector technique. The consistency ratio (CR) was calculated according to Wind and Saaty (1980) in order to evaluate the normalized weights of various thematic layers and their specific classes. The following steps were taken to compute the CR of various thematic layers and their classes:

Square matrix: A = \(a_{ij}\) [Eq. (7)], the element of row i column j was produced and the lower triangular matrix \(L_{ij}\) was completed by taking the reciprocal values of entries of the upper diagonal using the expression \(\left( {L_{ij} = \frac{1}{{a_{ij} }}} \right)\).

$$A = \left[ {\begin{array}{*{20}c} {\frac{{p_{1} }}{{p_{1} }}} & {\frac{{p_{1} }}{{p_{j} }}} & {\frac{{p_{1} }}{{p_{m} }}} \\ {\frac{{p_{i} }}{{p_{1} }}} & 1 & {\frac{{p_{i} }}{{p_{m} }}} \\ {\frac{{p_{n} }}{{p_{1} }}} & {\frac{{p_{n} }}{{p_{j} }}} & {\frac{{p_{n} }}{{p_{n} }}} \\ \end{array} } \right]$$
(7)

To get normalized relative weight, we divide each element of matrix \(a_{ij} = \frac{{p_{i} }}{{p_{j} }}\) by Eq. (7) to obtain Eq. (8):

$$\frac{{\frac{{p_{i} }}{{p_{j} }}}}{{\frac{{\mathop \sum \nolimits_{i = 1}^{n} p_{i} }}{{p_{i} }}}} = \frac{{p_{i} }}{{p_{j} }} \times \frac{{p_{j} }}{{\mathop \sum \nolimits_{i = 1}^{n} p_{i} }} = \frac{{p_{i} }}{{\mathop \sum \nolimits_{i = 1}^{n} p_{i} }}.$$
(8)

Averaging across the rows to get the normalized principal eigenvector (priority vector), i.e., Rate (Ri) or weight of row i (Wi) Eq. (9) resulted. Since it is normalized, the sum of all elements in the priority vector should be one.

$$\frac{{W_{i} }}{{R_{i} }} = \left[ {\frac{{P_{i} }}{{\mathop \sum \nolimits_{i = 1}^{n} P_{i} }} \ldots \ldots + \cdots \frac{{P_{i} }}{{\mathop \sum \nolimits_{i = 1}^{n} P_{1} }}} \right] \times {\raise0.7ex\hbox{$1$} \!\mathord{\left/ {\vphantom {1 n}}\right.\kern-\nulldelimiterspace} \!\lower0.7ex\hbox{$n$}}$$
(9)

To compute \(\lambda_{max}\), first multiply the normalized value by the respective weight and then, the values of the product are added together to get:

$$\lambda_{\max } = \frac{{C_{1} + C_{2} + C_{3} \ldots C_{n} }}{n},$$
(10)

where \(C_{1}\) to \(C_{n}\) are the \(\lambda\) value and n is the number of criteria. In this research, a \(\lambda_{\max }\) value of 5.1966 was obtained. Consistency index is a measure of consistency or degree of consistency of the judgment and computed using:

$$C_{i} = \frac{{\left( {\lambda_{\max } - n} \right)}}{n - 1},$$
(11)

where n is the number of criteria considered. A consistency index value of 0.04915 was obtained in this study. The judgments of the pairwise comparison within each thematic layer were checked by the consistency ratio (CR) (Wind and Saaty 1980). The value of CR was computed by Eq. (12):

$${\text{CR}} = \frac{{C_{i} }}{{{\text{RCI}}}},$$
(12)

where \(C_{i}\) is the consistency index, RCI is the Saaty’s ratio index which is computed by Saaty (1987), and for 5 variable, RCI value is 1.12. Computed CR obtained in this study is 0.04388. If the value of the CR is equal or less than 0.10, the inconsistency is acceptable. On the other hand, judgments should be revised if the CR is equal or greater than 0.10. The final map was achieved by overlaying all thematic maps using weighted overlay methods after assigning rates for classes in a layer and weights for thematic layers (Tamiru and Wagari 2021; Girma 2022). The groundwater potential index (GWPI) was estimated using the spatial analysis tool in ArcGIS 10.8 through Eq. (13):

$${\text{GWPI}} = \mathop \sum \limits_{i = 1}^{n} W_{i} \times R_{i},$$
(13)

where \(W_{i}\) is the weight for each thematic layer and \(R_{i}\) is the rates for the classes within a thematic layer derived from AHP.

Results and discussion

Vertical electrical sounding (VES) data and hydraulic parameters

The general statistics of the VES data results for the locations is presented in Table 1. The results provide information on the primary geo-electric indices (layer resistivities, thickness and depths). The input layers for 19 VES points have been used to assess the outcomes from hydro-geophysical parameters, which are shown in Table 3. The ground truthing constrained interpretation of VES data performed using manual and electronic procedures gave the VES curves with their geologic dependent characteristics given in Fig. 2. The constraint of interpretation with VES data ensured fidelity in the interpreted primary geo-electric indices (Uwa et al. 2019). The inferred results identified geological units with layers ranging from three to four layers. The results in Table 1 showcased the resistivity with the average of 512.4 \(\Omega {\text{m}}\) and range of 21.0 to 1532.6 \(\Omega {\text{m}}\) in the first layer. The second and third layers have resistivity ranges and mean values as 63.3–3688.7 \(\Omega {\text{m}}\) and 167.4–2984.5 \(\Omega {\text{m}}\) and 935.2 and 1112.6 \(\Omega {\text{m}}\), respectively. The fourth layers have average resistivity of 543.0 \(\Omega {\text{m}}\) and range of 61.6–1188.0 \(\Omega {\text{m}}\). In terms of thickness, layers 1–3 have ranges as 1.0–36.3 m, 8.0–247.7 m and 52.6–151.2 m, respectively, while their corresponding mean values are 7.2 m, 66.2 m and 93.2 m. The inverted depths of investigation for layers 1 to 3 have respective ranges of 1.0–36.3 m, 9.0–250.0 m and 63.9–165.4 m and their corresponding mean of 7.2 m, 37.8 m and 105.9 m. The values of thickness and depth of investigations were not defined in layer 4 at the maximum current electrode separations (Uwa et al. 2019; Thomas et al. 2020). The distributions of resistivities from top to bottom of depth of investigations indicate the presence of facies changes within the study area. From the available ground truth data, the topmost layer is motley in compositions and the subsurface is characterized by fine to gravely sands with negligible intercalations of argillites, a typical geologic composition of Benin Formation in the Niger Delta (Akpan et al. 2020a, b). The lithofacies changes delineated reflect clay \(\left( {\rho < 100 \, \Omega {\text{m}}} \right)\), sandy clay \(\left( {100 > \rho < 100 \, \Omega {\text{m}}} \right)\), fine sands,\(\left( {400 > \rho \le 1000 \, \Omega {\text{m}}} \right)\), medium grained sands \(\left( {1000 > \rho \le 1500 \, \Omega {\text{m}}} \right)\) and gravelly sands \(\left( {1500 \, \rho < 4000 \, \Omega {\text{m}}} \right)\). These high and low lithofacies indicate the variability in geologic units and hydraulic flow units in the study area (Umoh et al. 2022). The aquifer thickness reflects the volume of groundwater in the subsurface. Areas with the same thickness could be described in terms of isopach. Figure 4a and Table 3 reveal five major groups (50–92 m, 92–134 m, 134–177 m, 177–219 m, 219–262 m) of aquifer thicknesses delineated. The hotspot of the aquifer thickness clustered around VES 8, VES 14 and VES 15, which have the highest aquifer thickness in the range of 219–262 m. Aquifer depth refers to the location of groundwater in the subsurface. Areas with the same depth could be described in terms of Isopach. Figure 4b and Table 3 reveal five major groups: 51.48–94.13 m, 94.13–136.78 m, 136.78–179.43 m, 179.43–222.09 m and 222.09–264.74 m of aquifer depth inferred. The hotspot of the aquifer depth clustered around VES 9, VES 14 and VES 15, which have the highest aquifer depth in the range of 222.09–264.74 m. As all rocks have resistivities comparable to their nature and contents, rock units that make up aquifer resistivities are relatively high. The resistivities depend on the curve shapes of the VES data and model, which is dependent on geology of the place. This invariably means that there is no specific range of aquifer resistivities (George 2021a). The aquifer resistivities vary between 83.65 Ωm and as high as 3683.50 Ωm (Table 3 and Fig. 4c). It can be observed that the highest aquifer resistivities in this area range from 2963.53 to 3683.50 Ωm and the areas with this range include: VES 10 and VES 12. The ratio of empty space to total volume, which is commonly represented as fraction or in percentage, is the aquifer’s porosity (Fig. 4d and Table 3). Effective porosity, which measures the amount of linked void space to the total volume, is the space accessible for fluid to flow. How much water can be held in geological materials is determined by their porosity. Due to their porosity, groundwater is present in almost all rocks. VES 16 and VES 19 have the highest effective porosity and by implication permeability in the study area (Ekanem et al. 2022b).

Fig. 4
figure 4

Image map showing distribution a aquifer thickness, b aquifer depth, c aquifer resistivity, and d aquifer porosity distribution in the study area

Full size image

Based on the estimation of economic parametric properties, it is observed that hydraulic conductivity of the area is highest around the southeastern part of the study area where the greatest hydraulic conductivity is about 3.43 m/day around VES 1, 4, 16, 17 and 19 (Fig. 5a and Table 3). The hydraulic conductivity Kh values in the southeastern part of the study area are high, and this indicates aquifers of good potentials (Ekanem et al. 2022a). As hydraulic conductivity is a measure of the ease at which groundwater flows, it is directly related to the lithology of the location and again the resistivity and transmissivity of the place. High hydraulic conductivity is also a function of the depth or aquifer thickness. The values in Fig. 4a were derived from the locations where hydraulic heads of some places were measured and others were statistically extrapolated. It can be deduced for nearby locations or places of uniform geology where any measurement has not been done.

Fig. 5
figure 5

Image map showing distribution a hydraulic conductivity and b transmissivity distribution in the study area

Full size image

The transmissivity of the study area is shown in the map of Fig. 5b and Table 3. The values are in the range of 71–687 m2/day. The transmissivity increases northward of the study area, with the eastern part having the lowest values of transmissivity. The highest values of transmissivity are found in VES 9, VES 14 and VES 19.

The chart in Fig. 6 shows that there is little or no variation between the thickness and depth of the aquifers in the study area except in VES 3 and 15. It is also observed that deeper aquifers have larger thickness within the study area. This finding is in line with Ibuot et al. (2019) and George (2020).

Fig. 6
figure 6

Relationship between thickness and depth in the study area. There is little or no variation between the thickness and depth of the aquifers in the study area except in VES 3 and 15 as shown in Fig. 8. It is also observed that deeper aquifers have larger thickness within the study area

Full size image

Conversely, the aquifer bulk resistivity is highly variable within the study area with the highest value in VES 10. There exists no relationship between the transmissivity (the product of hydraulic conductivity and thickness) and bulk resistivity (Fig. 7) and hence the reason for the nonlinearity of transmissivity with resistivity in the study area. This shows that resistivity alone cannot be reliably used for groundwater potential prediction as obsessed by some drillers. Conductivity and thickness are very important in groundwater potential estimation. The findings of this study also identified the relationship between porosity (fractional) and hydraulic conductivity (m/day) in the study area. According to the chart in Fig. 8, there is a fairly invariant porosity distributions indicating that the study area is made up of a suite of sands with minimally varying pore spaces within them. However, hydraulic conductivity, which also depends on porosity, is controlled by dynamic viscosity and mean grain size (Soupios et al. 2007). In a nutshell, while fractional porosity is fairly constant due to similarity in lithology, hydraulic conductivity shows remarkable variations at different points due to varying dynamic viscosity and grain size. Practically, this shows the distribution of porosity and its hydraulic conductivity at different locations. This can be used to show the degree of permeability of the aquifer repositories (Umoh et al. 2022). As observed, high values of porosity correspond to high hydraulic conductivity. This finding agrees with George et al. (2018) and Uwa et al. (2019). Specifically, VES 18 and 19 have the highest hydraulic conductivity value even though porosity remains fairly constant. This further shows that porosity is fairly sensitive to the pore system, while hydraulic conductivity, a quantitative measure of the capability of a geologic formation or other porous media to transport a specific fluid, is highly sensitive to the pore system (Soupios et al. 2007).

Fig. 7
figure 7

Relationship between bulk resistivity and transmissivity in the study area

Full size image
Fig. 8
figure 8

Relationship between porosity (fractional) and hydraulic conductivity in the study area

Full size image

Based on the VES data analysis, Table 4 shows the groundwater potential and classification for the study area. From Table 4, VES stations 6 and11 have intermediate and withdrawal of local water supply (small community, etc.) according to Okogbue and Omonona (2013) classification. All other VES points in the study area have high classification and withdrawal of lesser regional importance according to the classification.

Table 4 Inferred classification and groundwater supply potential in the study area (Okogbue and Omonona 2013; Agbasi and Etuk 2016)
Full size table

RS and GIS analysis

Lineament and lineament density

Lineaments are linear features on the Earth’s surface that indicate groundwater fissures on the surface. They are characterized as secondary porosity and may be detected on satellite photographs as tone shifts when compared to other topographical features (Suganthi et al. 2013; Sahu et al. 2022). A lineament can be a crack, fault, fracture, or master joint, as well as long and linear geological structures, topographic linearity, or the straight route of a stream. The previous studies have discovered a connection between lineaments and groundwater storage and flow (Mabee et al. 1994; Magowe and Carr 1999; Solomon and Ghebreab 2006). The lineaments were imported into the ArcGIS environment, and the spatial analyst tool was used to build the lineament density map for the research region. Figure 9a depicts the lineament density map of the study area. A lineament density map is a quantitative length of linear feature per unit area that may indirectly reveal groundwater potential since the presence of lineaments typically indicates the existence of permeable units/layers. The line density tool was used to produce the lineament density from the lineament map of the research region. Higher lineament density polygons reflect more recharging and so have a better groundwater outlook (Solomon and Ghebreab 2006). The most important rating value was assigned to the largest lineament density period. Lineament junctions may also represent possible places of groundwater storage. As a result, places with high lineament density may have large groundwater possibilities. The map in Fig. 8a indicates uniformity in linear density (1 per kilometer) in all the VES points except VES 3, which has about 5 per kilometer.

Fig. 9
figure 9

Image map showing distribution of a reclassified lineament density, b land use/land cover, c drainage density and d slope in the study area

Full size image

Land use/land cover map

The occurrence of groundwater is influenced by the study area land cover and land use. Figure 9b illustrates the supervised classification of land use and land cover maps. According to the finding, three areas have the greatest area coverage in the research region. The built area is followed by water bodies and rangeland. Land usage and cover have an impact on runoff by either decreasing and assisting runoff or trapping water on plant leaves (Fan et al. 2022). Water droplets captured in this method fall to the ground and replenish groundwater. Land use/land cover may also have a detrimental impact on groundwater through evapotranspiration, assuming continual interception.

Drainage and drainage density maps

Figure 9C displays the research area drainage system. The drainage pattern was extracted straight from the DEM data. The type of vegetation, rainfall absorption capacity of soils, infiltration, and slope gradient have an impact on the drainage system, which is one of the most important indicators of hydrogeological features. The surface drainage density is defined as the ratio of total stream lengths to the grid area under consideration, and it is measured in per kilometer. A zone with a low drainage density promotes infiltration and decreases surface runoff (Cotton 1964; Luijendijk 2022). Generally, low drainage density areas are suitable for groundwater development. The analysis in Fig. 8c indicates that study area has a range of 1–5 km−1 of drainage density with maximum areal spread occupied by low density between 1 and 2 km−1 and hence indicating significant groundwater potential (Ekanem et al. 2022a).

Slope maps

One of the factors controlling groundwater infiltration and recharge is slope. Thus, the kind of slope, in conjunction with other geomorphic parameters, may indicate the potential of groundwater in a certain place. Surface runoff is minor in low slope locations, allowing more time for rainwater infiltration, whereas heavy runoff occurs in high slope areas, limiting residence time for infiltration and recharging. In this study, the slope map (Fig. 8d) of the region was created using DEM data and the spatial analyst tool in ArcGIS. The slope map in Fig. 9d is generally 1° (one degree). This indicates the possibility for high infiltration and recharge, which leads to enhanced water potential in the area (Magesh et al. 2012).

Soil type map

The soil type influences on both the porosity and permeability of aquifer rocks. The lithology layer was prepared by digitizing existing soil type map (Fig. 10a). The inferred soil type from RS is generally sandy with negligible sandy clay (Fig. 10a). This result from RS in Fig. 9a is also in line with the VES data results in Fig. 2. This suggests that the area is porous and permeable, and hence, it is characterized by high groundwater potential (Ekanem et al. 2022b).

Fig. 10
figure 10

Image map showing distribution of a soil type and b groundwater potential map of the study area

Full size image

Groundwater potential map

Integration of all thematic maps leads to the generation of groundwater potential index map in Fig. 10b. All the thematic maps of the elements influencing groundwater occurrence, transport, storage, and recharge were weighted and merged in an attempt to develop the basin's groundwater potential index (GPI) map. The image map in Fig. 10b depicts the RS generated map, and Table 5 shows the weights of the groundwater regulating elements. The image map shows four separate zones: poor, moderate, good and excellent. Furthermore, a thorough examination of the map indicates that the groundwater potential of the research region is mostly determined by lineament density, slope, and land use, in addition to soil type. The image also shows that places with significant groundwater potential are the north-western parts of the area having high lineament density and between low and average slope. In general, the area has moderate to good groundwater potential index. This assertion from the work is in agreement with the work by Ikpe et al. (2022) and Ekanem et al. (2022a, b).

Table 5 Weights of groundwater factors used in the study
Full size table

As inferred from analysis, the most important element, according to Table 5, was soil type (43%), followed by lineament density (33%), and slope (13%). Land use/land cover is the least major contributor to groundwater in the research region, accounting for about 4% of the total. The study CR determined to be 0.043884 is much lower than the 0.10 consistency threshold. In ArcGIS 10.8, the areal coverage of groundwater potential index was estimated by converting the potential to shapefiles and using the computation geometry tool in the characteristics table to construct each potential zone.

The analytical hierarchy process (AHP) was used in a GIS setting to merge five thematic maps of elements influencing groundwater occurrence and movement using weighted overlay. The geometry calculator was used to estimate the groundwater potential coverage. Finally, the possible groundwater zones were evaluated using geophysical survey sites from several of the research survey area selected. According to the findings, soil type, lineament density, slope, drainage density, and land use/land cover have impacted on groundwater possibilities in the study region (Fan et al. 2022). The weightages of the parameters that determine groundwater prospects revealed that soil type was weighed the most (43%), followed by lineament density (33%), slope (13%), and drainage density (7%). With a weight of 4%, land use/land cover is the least significant contributor to groundwater in the research region. The CR for this study was calculated to be 0.04, which is much below the 0.10 consistency standard. The groundwater potential in the area surveyed is largely moderate, with a high potential within the study limits. Low and extremely low groundwater potentials are typically found in the southern reaches.

The groundwater potential index has been determined and classified into four categories: excellent, good, moderate, and poor potential. The study found that high groundwater potential index is concentrated in the northwestern part of the study area around VES 3 and in the boundaries of the study area. This study, which is in analogy with Sahu et al. (2022), has found significant regional diversity in groundwater potential within the study area. The variability observed is strongly related to the variability in the lithology, slope, drainage density, lineament density, and land use/cover (Mogaji et al. 2021). The most attractive prospective zone has low to average slope and a high lineament density. The work has shown generally low slope (< \(2^\circ\)) and variable lineament density ranging from 1 to 5 \(km^{ - 1}\).

Validation of VES with RS and GIS

The validity of the result of RS and GIS technique was tested against the VES results. Cross-examination of the 19 VES geophysical surveyed points and the geospatial analyses was observed in the study. This was possible by overlaying the VES results on the groundwater potential map generated from the integration of various thematic maps estimated using multicriteria decision making. The attributes are summarized in Table 6 for the study area. The results showed maximally correlated and minimally infinitesimal uncorrelated relationships in the test groundwater potential parameters from VES data and the thematic geospatial data (RS and GIS). Out of the 19 points, only VES 1 and 11 points appeared to be grossly in disagreement in terms of groundwater potential among the electrical prospecting and spatiotemporal methods. This disagreement can be attributed to inherent departure of regional geology from local geology inferred from VES ground-based resistivity survey, vague multicriteria decision with deviation from normal or from the uniqueness in VES interpretation that is not screened by ground truth data (Adiat et al. 2012; Akpan et al. 2013). The result of overlay of the geophysical survey on the groundwater potential index obtained from the RS, GIS and VES technique in Table 5 showed reasonable seamless correlations in terms of groundwater potential with infinitesimal misfit expected to be due to obvious reason. The general results show that geospatially induced potential is in agreement with ground-based locally estimated potential and the hybridized map can be used seamlessly in groundwater exploration, exploitation or extraction.

Table 6 Groundwater potentiality from VES, RS and GIS in the study area
Full size table

Conclusion

The groundwater potential index has been derived in parts of Akwa Ibom State for the first time using contributions of electrical prospecting and spatiotemporal variations. The outcome showed that water repositories with high thickness and depths clustering around VES 9, VES 14, and VES 15 have the highest values ranging from 219 to 262 m and the highest aquifer depth in the range of 222.09–264.74 m. The highest aquifer resistivity in this area ranges from 2963 to 3683 Ωm. As inferred from analysis, the most important element was soil type (43%), followed by lineament density (33%), and slope (13%). Land use/land cover is the least major contributor to groundwater in the research region, accounting for about 4% of the total. The study CR determined to be 0.043884 is much lower than the 0.10 consistency threshold and hence the justification of the results. The comparison of local information from surface geophysical technique (VES) and regional/spatiotemporal information from GIS-based multicriteria decision making (MCDM) and AHP methods to produce groundwater-potential maps of the study area using logical groups of characterizing hydrogeological entities has shown a striking uniformity in the hydrogeological study involving the potential of groundwater repositories in the study area. Aquifer systems have been classified into four categories, namely excellent, good, moderate and poor potential. It was observed from the study that high groundwater potential index is located in the southwestern part of the study area. However, the study showed that there is a relationship between the vertical electrical sounding (VES) and remote sensing results as the results of overlay and interplay between them showed that areas with high groundwater potential from VES analysis align with areas of good groundwater potential index map produced by GIS and RS. This study opined that there is a large spatial variability of groundwater potential within the study area. The variability inferred is closely followed by the variability in the soil type, slope, drainage density, lineament density and land use/cover of the study area. The most promising potential zone in the area is related to low slope and high lineament density which agreed perfectly with Fan et al. (2022). This study also showed in general that the groundwater potentiality increases with an increase in the thickness of the aquifer and depth for both GIS and remote sensing techniques and also with the electric resistivity method. This result, which agreed with George (2020) ground-based geophysical technique, can be used as reliable inputs in accessing sustainable groundwater mostly during dry seasons that durable water-bearing units seem to be issues to be considered in accessing groundwater within the economic depth of the study area.