Received: 16 May 2023 Accepted: 26 August 2023
DOI: 10.1111/1365-2478.13422
O R I G I N A L A R T I C L E
Depth-based correlation analysis between the density of
lineaments in the crystalline basement’s weathered zones and
groundwater occurrences within the Voltaian basin, Ghana
Theophilus Yaw Amponsah1,2 David Dotse Wemegah2 Sylvester Kojo Danuor2
Eric Dominic Forson3,4
1Council for Scientific and Industrial Abstract
Research-Institute of Industrial Research,
Geological structures have been shown by studies to have influence on the occurrence,
Accra, Ghana
2 storage and transportation of groundwater. Understanding the structural network of anGeophysics Section, Department of
Physics, Kwame Nkrumah University of area unearths a deep insight into the groundwater dynamics of the area. A geologi-
Science and Technology, Kumasi, Ghana cal structural analysis was carried out to reveal the geological structural network of
3Department of Physics, School of Physical Ghana’s Voltaian basin. Using aeromagnetic data, structural density models were gen-
and Mathematical Sciences, University of
Ghana, Legon, Accra, Ghana erated using the Center for Exploration Targeting grid analysis technique for two depth
4CODES-ARC Centre of Excellence in Ore ranges (that is up to 100 m and 300 m) over the Voltaian basin. The total length of geo-
Deposits and School of Earth Science, logical structures (lineaments) delineated at depths up to 100 m and 300 m were more
University of Tasmania, Hobart, Tasmania,
than 5000 km and more than 8000 km, respectively. Given this, the study area was
Australia
observed to be structurally dense at each of the aforementioned depths. The structural
Correspondence density models were discretized into five classes (very low, low, moderate, moderately
Eric Dominic Forson, Department of
high and very high regions), each of which was evaluated to determine their spatial asso-
Physics, School of Physical and
Mathematical Sciences, University of ciation with known locations of groundwater occurrences within the study area using
Ghana, Legon, Accra, Ghana. Email: the frequency ratio technique. Frequency ratio results for both structural density models
ericdforson@gmail.com and
edforson@ug.edu.gh derived at 100 m and 300 m depths show the existence of a strong correlation between
high structural density model classes and the known groundwater occurrences. The
structural density models were further evaluated using the receiver operating character-
istics curve. The area under the receiver operating characteristics curve scores indicates
that, although both structural density models showed very good performance (with
receiver operating characteristics scores greater than 0.7), the 300-m depth structural
density model performed better than the structural density model generated at a depth
of 100 m (with their receiver operating characteristics scores being 0.721 and 0.715,
respectively). The obtained results corroborate with literature assertion that groundwa-
ter occurrence within the Voltaian basin is mainly associated with structural features. It
is expected that the outputs of this study would guide future groundwater exploration
programmes within the study area.
K E Y W O R D S
aeromagnetic data, frequency ratio, groundwater occurrence, lineaments, receiver operating character-
istics, structural density model, Voltaian basin
Geophysical Prospecting 2023;1–15. wileyonlinelibrary.com/journal/gpr © 2023 European Association of Geoscientists & Engineers. 1
2 YAW AMPONSAH ET AL.
INTRODUCTION enormous contribution to groundwater dynamics in the sub-
surface implies fractures are essential factors for groundwater
Surface water resources within the Voltaian basin are largely targeting and efficient exploration. Hydro-geophysical analy-
ephemeral owing to the several climatic conditions over sis, particularly in groundwater development related issues,
the region. The vulnerability of surface water resources to often includes aquifer potential evaluation, water quality
climatic variability explains the seasonal variations, which and contamination assessment as well as aquifer vulner-
result in water scarcity at certain periods of the year. The ability/protective capability analysis, among other things
climate-induced scarcity coupledwith the continuous destruc- (Kinnear et al., 2013; Nwachukwu et al., 2018).
tion of surface water resources has significantly resulted in In Ghana, vulnerability to drought are influenced by geo-
an increasing demand for groundwater as a potable water graphical patterns and affects socioeconomic conditions.
resource in many households and firms. This is very evi- According to national and regional vulnerability evaluations,
dent in the northern sector of Ghana where most surface the northern regions are the most vulnerable regions to
water resources are largely ephemeral due to harsh weather drought in Ghana. Also, these regions have the lowest adap-
conditions (Mul et al., 2015). Groundwater accounts for tation capability in the country due to poor socioeconomic
about 60% of the freshwater supplies in the Voltaian basin development and over-reliance on rain-fed systems like agri-
(Obuobie et al., 2016). Groundwater in a normal base- culture and forestry for local economies and livelihoods.
ment terrain is found in weathered and fractured aquifers as Climate change is negatively affecting rural livelihoods in
well as in the transition region (partly weathered basement) these places, as yearly rainfall is decreasing and rainfall pat-
between the weathered rock and the fresh bedrock (MacDon- terns are becoming more variable. Extreme weather events,
ald et al., 2011). In contrast to other sedimentary formations, drought and flooding are posing increasing problems to com-
aquifer lengths and depth extents differ greatly in the Volta- munities. Although water is a vital thematic objective in
ian basin (Mainoo et al., 2019). Groundwater availability Ghana’s development agenda, it is yet becoming increasingly
within the Voltaian basin is mainly due to secondary poros- scarce in northern Ghana due to climate change (Alhassan
ity, thus groundwater within the basin is strongly dependent & Hadwen, 2017). The study of the structural complexity is
on fractures and folds (Mul et al., 2015). hinged on the premise that groundwater accumulation, flow
Fractures and folds are important structural characteristics and storage are largely dependent on the geological structures
that influence groundwater. Fractures are created by the brit- in the Voltaian basin (Mul et al., 2015). This is reported to
tle breaking of rocks and are classified as joints, fissures and be the case because groundwater in the Voltaian basin mainly
faults (Roberts, 1982). Folds are formed by ductile defor- occurs and flows in fracture zones and along bedding planes
mation, and the magnitude of the features formed, including for some areas due to the fact that the primary porosity of
synclines and anticlines, is determined by the degree of this these rocks is destroyed through consolidation and cemen-
deformation. Fractures function as hydrodynamic conduits tation (Loh et al., 2020). Structural complexity delineation
and hence are essential pathfinders to groundwater aquifer’s therefore provides insight into groundwater availability and
flow pattern. They could serve as hydraulic conductors or as accessibility (Yidana, 2010). A critical part of the structural
flow barriers in shallow and deep flow. In geological terranes complexity analysis is the depth at which these structures
such as the Voltaian basin, where primary porosity and per- occur as this provides information about the accessibility of
meability are absent owing to cementation and sedimentation, groundwater in the area as well as the operational depth for
the occurrence of structures is the most important feature that groundwater explorers. The method of exploration is there-
facilitates groundwater recharge and flow in the subsurface after based on this premise. The occurrence of deep-seated
(Mul et al., 2015). The detection of various zones of geo- water-bearing structures implies that groundwater sources in
logical structure occurrence, which also include fault zones, the area are deep seated, and hence groundwater exploration
fold hinges and joints are essential components that guide methods capable of probing deeper must be employed to reach
the selection of target areas for groundwater extraction. In these depths. However, if the structures are found to be within
view of this, Sankar (2002) suggested that sites of boreholes the near subsurface (thus structures are shallow seated), then a
should be selected with careful consideration of the litho- shallow groundwater exploration techniquemust be employed
logical and structural environment. While the significance of (Kinnear et al., 2013; Nwachukwu et al., 2018). It is also
geological structures (lineaments) is widely acknowledged, worth noting that the assessment of structural density (struc-
further studies are needed to fully comprehend the function tural occurrence) at deeper depths and their relationship with
and behaviour of these features. Thus, areas with high struc- groundwater over the Voltaian basin has been rarely studied.
tural density can have significant groundwater prospects even Thus, this study seeks to complement efforts in groundwater
in hilly regions. Geological structural network mapping is research in the Voltaian basin by assessing the relationship
regarded as a critical method used in detecting hot springs between the groundwater potential of the various geological
and conducting hydrogeological research (Sabins, 1999). Its structural networks of the area using the frequency ratio and
 13652478, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1365-2478.13422 by University of Ghana - Accra, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
DEPTH-BASED CORRELATION ANALYSIS 3
F I G U R E 1 A map showing various administrative regions within the study area.
the receiver operating characteristics techniques. It would also of the entire Voltaian basin, as shown in Figure. 1, which
provide information on the depth of occurrence of geologi- has also been known to occupy about one-third of the total
cal structures in the region as well as their spatial correlation land surface area of Ghana. The study area covers areas in
with groundwater occurrence within the region. The find- the Northern region and some significant parts of the Upper
ings of this research would help in efforts aimed at making East, Brong Ahafo, Savannah and Oti regions of Ghana.
water available and accessible. This would provide the needed Some major towns within the study area include Pusiga,
boost in the socio-economics of the area and would subse- Daboya, Buipe, Tolon, Savelegu, Dabogshe, Nyampala,
quently push the region a step further in the pursuit of poverty Tong, Damango, Yapei, Yendi and Bimbila among others.
alleviation and accessibility to potable water as envisaged in
the United Nations Sustainable Development Goals 1 and 6
(SDG1 and SDG6). Geology and hydrogeology of the Voltaian basin
The area within which this study is undertaken (Figure 1)
STUDY AREA covers the northern portion of the Voltaian basin, which has
been classified into three major lithostratigraphic schemes
Location and climate (Figure 2) (Carney et al., 2010). The Voltaian basin is pri-
marily composed of sediments that lie flat or gradually dip
The study area is located within latitudes 0.369◦ E and and sit uncomformably on the predominant Precambrian sys-
1.971◦ W and longitudes 10.788◦ N and 8.469◦ N in the tem (Griffis et al., 2002). The unconformities are caused by
northern part of Ghana (Service, 2014). It covers a total land the high rate of erosion that spans a considerable portion of
surface area of about 44,558.6 km2, making up about half the basin (Griffis et al., 2002). It is a heterogeneous basin that
 13652478, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1365-2478.13422 by University of Ghana - Accra, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4 YAW AMPONSAH ET AL.
F I G U R E 2 Geological map of the study area (modified after GSD, 2010.).
is primarily made up of limestone, mudstones, shale, sand- porosity, hydrogeological conditions are determined by the
stones and sandy and pebbly beds. This basin’s inferred age existence and intensity of weathering and fracturing (Yidana
and correlation are related to the Neoproterozoic and Late et al., 2019). The Voltaian basin has a fairly limited ground-
Meso-Proterozoic epochs (Jordan et al., 2009). Lithostrati- water capacity in general, though some water does come from
graphically, the Voltaian basin is classified into the Obosum fractures in the clay or loose portions of the arenaceous group
Group, Oti-Pendjari Group and Kwahu Group, which are gen- (Chegbeleh et al., 2009). Regional hydrogeology surveys have
erally referred to as the Upper Voltaian, Middle Voltaian, and revealed that cracks or joints are irregular in some regions
Lower Voltaian, respectively, in the old literature (Annan- and frequently absent. The hydrogeology of the area has been
Yorke & Cudjoe, 1971). The study area is underlain by rocks well explained by T. Y. Amponsah et al. (2022). This shows
from the Sang Conglomerate, Tamale Sandstone Formation that the capacity of groundwater in the region is complex and
and undivided Obosom Formation (Obosom Group) as well demands rigorous research strategies to unravel its prospects
as Afram Formation, Bimbilla Formation, Buipe Limestone and dynamics. The majority of groundwater found in this ter-
Member, Bunya Formation, Chereponi Sandstone Member rane is hosted in semi-confined and confined aquifers. A good
and Darebe Tuff Member of the Oti-Pendjari Group. The area number of hydrogeologists, such as Kortatsi (1994), have doc-
also features the Kwahu Group constituted by the Anyaboni umented that the yield potential is generally poor in Voltaian
Sandstone Formation, Damongo Formation, Panabako Sand- rocks. The hydraulic characteristics of the Voltaian sedimen-
stone Formation and Tossiegou Formation (as shown in tary environment were also characterized as heterogeneous
Figure 2) (Jordan et al., 2009). Except in a few places, the with themoderate-to-low transmissivity values ranging exten-
Voltaian rocks are mainly flat and impermeable. There may be sively from 0.2 to 267 m2 per day in the clay-rich Obosum
exceptions in locations where due to weathering porous and shale and mudstone environment (Dzigbodi-Adjimah, 1993).
permeable surface materials have formed from jointed sand- Available data reveal that the mean depth of boreholes for
stones, arkoses and quartzite. Because the rock lacks primary most areas is around 90 m and the average depth is 48.1 m
 13652478, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1365-2478.13422 by University of Ghana - Accra, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
DEPTH-BASED CORRELATION ANALYSIS 5
T A B L E 1 Survey parameters and specifications for the airborne Data processing
magnetic survey (Jordan et al., 2009).
Survey parameter Specification used The magnetic method has progressed from its initial use as a
tool for finding iron ore to a common tool for mineral, hydro-
Minimum line length 1200 m
carbon, groundwater and geothermal resource exploration.
Flight line spacing 500 m
The technique is often widely used in other applications,
Trend of the flight line 135˚
such as water-resource assessment studies and determining
Tie line spacing 5000 m
the best location for water deposition based on the depth of
Trend of the tie line 225˚
the basement and other related geological structures. It is also
Nominal terrain clearance 75 m used to assess the thickness of surface alluvial sediments and
Data sampling time 0.05 s their basins (Blakely et al., 1973; Smith et al., 2004). In this
study, the use of aeromagnetic data played a primal role in
the delineation of various geological structures (lineaments)
(Yidana et al., 2019). However, in the far-eastern part of the as well as their densities over the study area. This was pre-
system, there is a record of just a few boreholes above 100 ceded by applying the reduction-to-equator (RTE) technique
m, even to around 150 m. The weathered or loosened areas to the residual magnetic intensity (shown in Figure 3) and sub-
range from 4 to 20 m in some parts of the south side of the sequently inverting it to generate magnetic responses that are
Voltaian basin, and inhabitants rely on the hand-dug boreholes symmetrical to their respective sources, which is denoted as
for their water supplies (Chegbeleh et al., 2009). Boreholes RTE_Inv. Afterwards, the first vertical derivative as an edge
yield values ranging between 5 and 1200 L/min, the amount of detection method was employed on the RTE_Inv to delineate
static water is 1–20 m, and the fluctuations in the water table various structures within the study area. The upward contin-
average about 4 m (Acheampong & Hess, 1998). Transmis- uation and downward continuation (DC) filtering techniques
sivities are measured between 0.3 and 270 m2/day (Darko & were, respectively, applied on the RTE_Inv grid to further
Krásny, 2007). Groundwater recharge depends on the surface transform and visualize the magnetic intensity responses at
infiltration and permeability of the nearby geological mate- different depths below the surface of the Earth. In the case of
rial that shields the aquifer from one location to another. The the upward continuation, it was carried out on the RTE_Inv
available literature shows that the annual precipitation within to suppress shallow anomalies hence enhancing deep-seated
the Voltaian basin ranges between 3.7% and 5% (Apambire, anomalies at a depth of 300 m. Thus, the upward continuation
2000; Martin & Van De Giesen, 2005). The annual ground- (UC) operation filter aided the transformation of the potential
water extraction is estimated at less than 5% (Martin & Van field (aeromagnetic) data measured on one surface to a higher
De Giesen, 2005). surface (Nabighian et al., 2005). The use of the UC filter was
plausible because it preserves primarily longer wavelengths at
higher observation levels, whereas short-wavelength anoma-
METHODS lies caused by small-sized near-surface structures and/or the
effect of cultural features (Roest et al., 1992). For the DC of
Data the RTE_Inv grid, its usage led to the delineation of short
wavelength magnetic anomalies up to a depth of 100 m from
The main data used in this study were sourced from an air- the Earth’s surface. The process was vital in the optimiza-
borne magnetic survey that was carried out in 2008 over the tion of the magnetic response of formations and structures
entire Keta and Volta basins in Ghana by Fugro Airborne Sur- in the top 100 m of the crust, where groundwater reserves
veys of Perth under the Mining Sector Support Program of are concentrated. Because the shortest wavelengths and data
the Government of Ghana and supervised by the Ghana Geo- noise are amplified exponentially, the DC was employed with
logical Survey Authority (GSD, 2010; Jordan et al., 2009). extreme caution (Reeves, 1989). All these aforementioned
The magnetic data were processed and analysed further to processing procedures were carried out using the Geosoft
increase their quality and render them more applicable to the Oasis Montaj software.
research objectives. The data was taken through various stages
of enhancement to improve its quality. This made it easier to
delineate fault, fracture and contact features that may regulate Delineation of structures and structural complexity
groundwater dynamics such as accumulation and flow in the analysis
study area (Reeves, 1989). The survey parameters and spec-
ifications employed during the airborne magnetic survey are Variations in the crystalline basement and volcanic debris
summarized in Table 1. are the primary causes of magnetic anomalies. Sediments
 13652478, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1365-2478.13422 by University of Ghana - Accra, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
6 YAW AMPONSAH ET AL.
F I G U R E 3 Map of the residual magnetic intensity .
are often magnetically “transparent,” implying that their sus- ysed to compute the local variations within the magnetic
ceptibility is negligible. As a result, interpreting magnetic responses observed at each continued level; (2) various lin-
data can be a useful tool for investigating and analysing a eations (linear features) were identified based on the phase
basin’s deepest and most fundamental geological structures. symmetry approach; (3) amplitude thresholding was carried
According to Kheyrollahi et al. (2016), one of the most impor- out to find ridges, followed by line thinning (skeletonization)
tant goals of the use of magnetic data is the delineation of procedure and then vectorizing the skeletonized structures
geological formations and structures. Generally, groundwa- (thinned lines) to generate a vector lineament data (shown in
ter accumulation, flow and storage in the Voltaian basin are Figure 7a) that would subsequently serve as input for the gen-
largely controlled by the geological structures such as frac- eration of the structural density heat map, otherwise known
ture zones along bedding planes for some areas. This is due to as the structural density map (Holden et al., 2012).
the fact that the primary porosity of these rocks is destroyed
through consolidation and cementation. The geology, on the
other hand, is based on the bedrock lithology as well as the Evaluation and validation of structural density
regolith. The regolith is reported to be unsaturated in many models
areas and would thus only provide minor amounts of ground-
water locally (Acheampong &Hess, 1998). In order to outline In this study, the structural density models generated for depth
the structural network of the area, the first vertical deriva- ranges 0–100 m and 100–300 m were evaluated and validated
tive filter was employed on the RTE_Inv grid to delineate using the frequency ratio approach and the receiver operat-
various lineaments. ing characteristics (ROC) technique. The frequency ratio (FR)
The CET (Center of Exploration Targeting) grid analysis is a bivariate geostatistical approach that determines the spa-
technique in Geosoft Oasis Montaj was applied to the down- tial correlation of various classes of a geospatial model with
wardly continued and upwardly continued RTE_Inv grids to respect to known locations of natural resource occurrence
obtain structural lineaments up to 100 and 300 m, respec- over a region of study. Thus, the aforementioned approach
tively. In carrying out this CET grid analysis technique, (1) has proven worthy in determining the spatial association of
each of the continued RTE_Inv grids was textually anal- various geospatial models with respect to known locations
 13652478, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1365-2478.13422 by University of Ghana - Accra, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
DEPTH-BASED CORRELATION ANALYSIS 7
of occurrences such as groundwater (P. O. Amponsah et al., However, the northern, western and southwestern parts of the
2023), mineral (P. O. Amponsah & Forson, 2023; Ghezelbash study area were delineated to be highly fractured. These clus-
et al., 2021), flooding (Forson, Amponsah, et al., 2023; For- ters of geological structures among other factors play a major
son & Menyeh, 2023), hydrocarbon (Arab Amiri et al., 2015) role in the groundwater dynamics of these areas. Hence, the
and landslide (Vakhshoori & Zare, 2016). The FR score for delineated geological structural network (lineaments) acts
a particular class 𝑖 (𝐹𝑅𝑖) of the structural density model is as underground water channels (conduits for groundwater
mathematically expressed as shown in Equation (1). occurrence). This fault network has the potential to turn
rocks into excellent aquifers. Again, faults act as drains,
𝐹BH BH= = 𝑖
∕BH𝑇
FR𝑖 , (1)
which influence the water level and thus affect groundwater
𝐹DM DM𝑖∕DM𝑇 distribution over an area of interest (Mulwa et al., 2009).
where 𝐹BH in Equation (1) depicts the groundwater occur-
rences (borehole points) within the class 𝑖 of the structural General structural characteristics (complexity)
density model; 𝐹DM is the areal frequency of class 𝑖 within of the study area
the structural density model; BH𝑖 is the number of boreholes
in class 𝑖 of the structural density model generated; BH𝑇 is the Figure 5 characterizes various lineaments delineated by
total number of boreholes over the study area; DM𝑖 is the area employing the Center of Exploration Targeting (CET) grid
of class 𝑖 in a generated structural density model, and DM𝑇 is analysis technique on the RTE_Inv (shown in Figure 3b)
the total area of the structural density produced. over the study area. These lineaments depict aquifer fractures
For a selected class 𝑖, the FR value greater than 1 depicts a that are capable of enhancing the permeability and porosity
strong correlation between the selected class and known loca- within the area. Thus, lineaments to a great extent influ-
tions of groundwater occurrence (borehole points) whereas ence the occurrence and storage of groundwater within areas
an FR value less than 1 indicates that the selected class is of their occurrence. The nature of the fractures can be an
weakly associated with the known locations of groundwater important factor in the morphology of the fractures (Acheam-
occurrences (borehole points). pong & Hess, 1998). The different colouration observed
The structural densitymodels obtained for each depth range from the legend of Figure 5a) depicts the trending direc-
(0–100 m and 100–300 m) were further evaluated by overlay- tions of the delineated structures over the study area. In
ing borehole data with depths within their respective levels view of this, eastern trending structures have been indi-
onto the structural density map (at depths 100 and 300 m) cated by deep green colouration whereas the structures which
to determine the relationship between highly fractured zones trend in the northern direction are represented by light green
and available borehole data. The efficacy of the relationship colours. North-western trending structures are depicted by
between the delineated structural density and borehole data blue colouration. Southern trending structures are shown in
was validated using the area under the receiver operating char- brown colours, and south-eastern trending structures are indi-
acteristics (ROC-AUC) curve. Area under the ROC values cated in pink colours. The rose diagram (Figure 5b) shows
range from 0.5 to 1, with values close to 1 indicating out- the trend of lineaments in the study area. It is observed that
standing performance and values close to 0.5 indicating poor the majority of the lineaments delineated within the study
prediction accuracy (P. O. Amponsah et al., 2023). trend in the eastern direction. A substantial amount of the
delineated structures also trend in the north-eastern and south-
RESULTS eastern directions. The area is dominated by eastern trending
structures with a population of 1493 (50.61%). Dense regions
First vertical derivative map of structural lineaments are essential to groundwater delin-
eation because they depict low-pressure zones where fluids
Figure 4 depicts the first vertical derivative (1VD) of the mag- such as groundwater can settle (Forson et al., 2020; Forson,
netic responses within the study area. An essential component Wemegah, et al., 2022). The occurrence of structures within
of the use of the 1VD technique is that it has the capacity to the Voltaian basin as established in the literature is a good
enhance anomalies associated with various structural features indicator of potential groundwater zones since groundwater
over a region of interest (Forson et al., 2021). In view of this, movement and storage are largely dependent on the structures
the generation of the 1VDmap over the study area was vital in in the area (Yidana et al., 2019). Figure 6 depicts the structural
the delineation of various geological structures over the study density over the study area and outlines highly, moderately
area with the northern, mid-western and southern clusters as and lowly fractured zones represented in red, green and blue
highly fractured. colours, respectively. From Figure 6, the northern and the
From Figure 4, the central and eastern portions of the western parts of the study area were generally delineated as
study area were, however, delineated to be poorly fractured. highly dense in terms of fracturing. Within the northern and
 13652478, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1365-2478.13422 by University of Ghana - Accra, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
8 YAW AMPONSAH ET AL.
F I G U R E 4 Qualitative image depicting responses generated based on the first vertical derivative of the magnetic field in a grey scale.
western regions, are characterised by acronyms comprising mostly within the first 100 m beneath the Earth’s surface with
AN, OB, PA and CHwhich are dominantly composed of sand- a few boreholes going beyond the 100-m depth (Agyekum
stone formation as well as BM, which is predominantly made & Asare, 2016; Manu et al., 2019). Various groundwater
up of mudstones. The area marked as BY, which is domi- explorers rely on drilling outcomes within these shallow
nated by sandstones, was also delineated to bemoderate to low depth ranges (0–100 m) to assess the groundwater prospects
dense zones in terms of structures. Groundwater availability is of the Voltaian basin. It is worth noting that the capacity
more likely to be plausible in areas with high structural den- of the fracture dynamics (structural occurrences) at deeper
sity, whereas regions with low structural densities generally depths within the basin has been rarely assessed in the
depict regions of low groundwater availability (Carney et al., literature (Carrier et al., 2008). In view of this, an under-
2010). standing of the groundwater-producible zones of this basin
requires knowledge of the depth of occurrence of geolog-
ical structures in the basin. Figures 7a and 7b represents
Structural complexity analysis from the Earth’s the delineated structural lineaments at depths 100 and 300
surface to 100 m and from 100 m to 300 m over m, respectively. The minimum length of the delineated geo-
the study area logical structures at the 100-m depth is 565.685 m (0.5657
km), whereas the maximum length of structures delineated
Producible depth ranges for groundwater exploration in at the same depth is 26,273.942 m (26.274 km). Up to a
Ghana especially within the northern sector of Ghana are depth of 100 m, the total length of delineated geological
 13652478, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1365-2478.13422 by University of Ghana - Accra, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
DEPTH-BASED CORRELATION ANALYSIS 9
F I G U R E 5 (a) Generalized structural complexity of the study area; (b) rose diagram depicting the trending direction of variously delineated
structures within the study area.
F I G U R E 6 Generalized structural density map over the Voltaian basin.
 13652478, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1365-2478.13422 by University of Ghana - Accra, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
10 YAW AMPONSAH ET AL.
F I G U R E 7 Delineated structures up to a depth of (a) 100 m and (b) 300 m within the Earth.
structures was 5,555,172.063 m (5555.172 km). The map The structural density models shown in Figure 8a,b charac-
(shown in Figure 7a) shows the northern portions to be more terize regions of high, moderate and low structural complexi-
fractured, while the central and south-eastern portions are ties (depicted as red, green and blue) observed at depths 100m
less fractured based on the structures delineated within the and 300 m over the study area. It can be observed that regions
first 100 m of the Earth’s surface. For depths up to 300 which were delineated to be predominant with the delineated
m within the Earth, the minimum and maximum lengths of structures (shown in Figure 7a,b) were also delineated to be
all the geological structures delineated were, respectively, highly dense in terms of structures on the structural density
447.2 and 27,103.5 m. The total length of geological struc- maps (Figure 8a and b). The range of structural density val-
tures outlined at the 300-m depth was 8,645,461 m. The ues obtained for various structures delineated up to a depth
map (shown in Figure 7b) shows several lineaments delin- of 100 m was from 0 (very low structural density zones) to
eated over the northern, western and southwestern parts of 110.9 km/km2 (high structural density zones). These highly
the study area. This indicates the occurrence of intense frac- dense or intense regions of structural occurrences characterize
turing within those regions. Although the central portions highly fractured zones and are generally expected that ground-
and south-eastern portions are less fractured in terms of the water potential in these areas would be significantly higher
300-m depth-delineated lineaments (Figure 7b), the few delin- (Preeja et al., 2011). For the structures delineated at a depth
eated structures in these areas are more dense and prominent of 300 m, the value of the density or intensity of various struc-
than those delineated within the same zones observed on the tures delineated ranged from 0 for very low-intensity regions
100-m depth structural map (shown in Figure 7a). In gen- to 158.1 km/km2 for regions of high structural density. It can
eral, the delineated structures at the 300-m depth are more be observed that regions delineated to be characterized by
dense and longer than those delineated at the 100-m depth. moderate-to-high structural density occurrence at the 300-m
The implication of this comparison, therefore, asserts that the derived SDM (Figure 8b) were more than that observed on the
lineaments or structures delineated at a depth of 300 m are 100-m derived SDM (Figure 8a). Since high structural density
more plausible for delineating groundwater within the study zones depict favourable zones of groundwater occurrence, it,
area than at a depth of 100 m since areas of intense fractur- therefore, suggests that groundwater prospects are higher at
ing (densely populated structures) depict low-pressure zones deeper depths (up to a depth of 300 m) than in the shallower
viable for hosting fluids including water (Forson et al., 2020; depths (up to a depth of 100 m) (Srivastava & Bhattacharya,
Nsiah et al., 2018). 2006).
 13652478, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1365-2478.13422 by University of Ghana - Accra, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
DEPTH-BASED CORRELATION ANALYSIS 11
F I G U R E 8 Structural density model up to a depth of (a) 100 m and (b) 300 m within the Earth.
T A B L E 2 Discretized classes of structural density models and borehole yields (groundwater occurrences).
Class description SDM class values (km/km𝟐) Classes of borehole yields (m𝟑/s) Class score
Very low ≤31.0 ≤10.0 1
Low 31.1–59.9 10.1–40.0 2
100-m depth- Moderate 60.0–80.0 40.1–150.0 3
derived SDM Moderately high 80.1–100.0 150.1–500.0 4
Very high >100.0 >500.0 5
Very low ≤32.0 ≤10.0 1
Low 32.1–62.0 10.1–40.0 2
300-m depth- Moderate 62.1–95.0 40.1–150.0 3
derived SDM Moderately high 95.1–130.0 150.1–500.0 4
Very high >130.0 >500.0 5
Abbreviation: SDM, Structural density model.
Evaluation of structural density models using technique was essential in determining the spatial association
the frequency ratio approach of each class of the SDM with respect to known locations
of groundwater occurrences within it. A summary of the fre-
In order to assess the spatial relationship between the SDM quency ratio results obtained for each class of the two SDM
produced and groundwater occurrence locations within them, generated is presented in Table 3. In the case of the SDM gen-
they were discretized into five classes of very low, low, erated at a depth of 100 m, the very low SDM class shows
moderate, moderately high and very high structural density a weak correlation with the groundwater occurrences within
zones based on the Jenks natural breaking classification tech- the model due to the 0.744 frequency ratio (FR) value (less
nique (Forson, Amponsah, et al., 2023; Jenks, 1963). This than 1) obtained. Groundwater occurrences were observed
was carried out by overlaying known locations of ground- to exhibit strong coherence with respect to the four other
water occurrences (borehole yields) on the SDM (shown in SDM classes, and thus the frequency ratio values obtained are
Figure 9a,b). The description of the discretization results is all greater than 1 (shown in Table 3). However, the moder-
summarized in Table 2. The application of the frequency ratio ately high class of the 100-m depth-generated SDM exhibited
 13652478, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1365-2478.13422 by University of Ghana - Accra, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
12 YAW AMPONSAH ET AL.
F I G U R E 9 Discretized SDM derived (a) at 100 m and (b) at 300 m depth overlaid with corresponding known locations of groundwater.
T A B L E 3 Frequency ratio values obtained for various classes within each of the two structural density models generated.
Structural density model (SDM) Class SDM class frequency (𝑭𝐃𝐌) Groundwater occurrence frequency (𝑭𝐁𝐇) Frequency ratio
Very low 0.661 0.492 0.744
Up to a depth Low 0.200 0.290 1.450
of 100 m Moderate 0.093 0.140 1.505
Moderately high 0.037 0.070 1.892
Very high 0.008 0.009 1.125
Very low 0.509 0.321 0.631
Up to a depth Low 0.218 0.179 0.821
of 300 m Moderate 0.1606 0.286 1.781
Moderately high 0.081 0.143 1.765
Very high 0.032 0.071 2.219
the strongest spatial relationship to groundwater occurrence. associated with fractured zones (structurally dense regions)
For the SDM produced for various lineaments delineated at (Loh et al., 2020). However, the high structural density class
a depth of 300 m within the Earth, the very low and low within the 300-m depth-derived SDM was observed to be the
SDM classes were observed to show a weak association with class with the strongest association with groundwater occur-
groundwater occurrence with FR values of 0.631 and 0.821, rence among all the classes within the two SDM-generated
respectively. The moderate, moderately high and very high (SDM at 100 m and 300 m) data.
classes of the 300-m depth-generated SDM exhibited a strong
correlation to groundwater occurrences within the study area
at that depth with FR values of 1.781, 1.765 and 2.219, Validation of structural density models using
respectively. In both structural density models, classes of high the receiver operating characteristics curve
structural density were observed to have frequency ratio val-
ues greater than 1. This further indicates that groundwater The spatial correlation between classes of SDM models
occurrence or prospect within the study area is strongly asso- and borehole yields (known locations of groundwater occur-
ciated with regions of high structural density. The frequency rence) was further assessed and validated using the receiver
ratio results obtained, therefore, corroborate with the litera- operating characteristics (ROC) technique. For the 100-m
ture that groundwater prospects within the Voltaian basin are depth-derived SDM, an ROC output (shown in Figure 10) was
 13652478, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1365-2478.13422 by University of Ghana - Accra, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
DEPTH-BASED CORRELATION ANALYSIS 13
F I G U R E 1 0 Area under the ROC curve
score for the SDM at 100 m and 300 m.
produced by employing 378 operational boreholes with yields (SDM) generated at a depth of 100 m and 300 m were eval-
ranging from below 10 to above 500 m3/s as validation point uated to ascertain their spatial association with respect to
data. Thirty eight (38) operational boreholes were used for the known locations of groundwater occurrences (borehole
validating the 300-m depth-derived SDM based on the ROC yields) using the frequency ratio technique. The frequency
results also shown in Figure 10. The outputs generated for ratio results obtained for low structural density classes indi-
the ROC were produced by discretizing the borehole yields cate a weak correlation with respect to the groundwater
into five classes depicting very low, low, moderate, moder- occurrences whereas the high structural density classes exhib-
ately high and very high yields of groundwater as shown in ited strong spatial association with respect to the known
Table 2. The score obtained for the area under the ROC curve locations of groundwater occurrences. The frequency ratio
(Area under the ROC [AUC]) for SDM derived at a depth of results therefore corroborate with the literature that ground-
100 and 300 m was, respectively, 0.715 (71.5%) and 0.721 water prospects within the Voltaian basin are associated with
(72.1%). The AUC score obtained (both of which are greater fractured zones (structurally dense regions). To further evalu-
than 0.7) for each of the structural density models indicates ate and validate the structural density models with respect to
that the models are efficiently produced. Nevertheless, com- groundwater occurrences, the area under the receiver operat-
paring the AUC score obtained for the 100-m depth-derived ing characteristics (ROC) technique was employed. Results
SDM derived to the 300-m depth-derived SDM indicates a obtained based on the ROC curve generated for the SDM
more efficient correlation between various classes of ground- derived at a depth of 100 m and 300 m were, respectively,
water occurrence with respect to classes of SDM at a depth of 0.715 and 0.721. Although the efficacy of the two SDM-
300 m (i.e., SDM derived at 300 m is more efficient than the generated data was deemed very good (since ROC scores are
SDM derived at a depth of 100 m). greater than 0.7), the SDM derived at a depth of 300 m was
observed to have a better performance than the SDM derived
at a depth of 100 m.
CONCLUSION
A C K N O W L E D G E M E N T S
In order to examine the relationship between groundwater The authors wish to thank the Ghana Geological Survey
occurrences and geological structures over the Voltaian basin, Authority, Accra, and the Community Water and Sanitation
this study was carried out to delineate the density of var- Agency of the Northern Region of Ghana for making the
ious geological structures (structural density models) at a data available. Many thanks to the University of Ghana–
depth of 100 m and 300 m over the study area using the Carnegie Corporation and the Building a New Generation
continuation techniques and the Centre of Exploration Tar- Africa (BaNGA-Africa) for their immense support in making
geting grid analysis technique. The structural density models this study a success.
 13652478, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1365-2478.13422 by University of Ghana - Accra, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
14 YAW AMPONSAH ET AL.
D AT A AVA I L A B I L I T Y S T AT E M E N T Forson, E.D., Amponsah, P.O., Hagan, G.B. & Sapah, M.S. (2023) Fre-
Data are available upon request to the corresponding author. quency ratio-based flood vulnerability modeling over the Greater
Accra Region of Ghana. Modeling Earth Systems and Environment,
R E F E R E N C E S 9(2),2081–2100.
Forson, E.D. & Menyeh, A. (2023) Best worst method-based mineral
Acheampong, S.Y. & Hess, J.W. (1998) Hydrogeologic and hydrochem- prospectivity modeling over the Central part of the Southern Kibi–
ical framework of the shallow groundwater system in the southern Winneba belt of Ghana. Earth Science Informatics, 16, 1657–1676.
Voltaian Sedimentary Basin, Ghana. Hydrogeology Journal, 6(4), Forson, E.D., Menyeh, A. & Wemegah, D.D. (2021) Mapping litholog-
527–537. ical units, structural lineaments and alteration zones in the Southern
Agyekum, W. & Asare, E. (2016) Challenges associated with ground- Kibi-Winneba belt of Ghana using integrated geophysical and remote
water resources development in northern Ghana. Ghana Journal of sensing datasets. Ore Geology Reviews, 137, 104271.
Science, 56, 39–51. Forson, E.D., Menyeh, A., Wemegah, D.D., Danuor, S.K., Adjovu,
Alhassan, S. & Hadwen, W.L. (2017) Challenges and opportunities I. & Appiah, I. (2020) Mesothermal gold prospectivity mapping
for mainstreaming climate change adaptation into wash development of the southern Kibi– Winneba belt of Ghana based on fuzzy
planning in Ghana. International Journal of Environmental Research analytical hierarchy process, concentration-area (CA) fractal model
and Public Health, 14(7), 749. and prediction-area (PA) plot. Journal of Applied Geophysics, 174,
Amponsah, P.O. & Forson, E.D. (2023) Geospatial modelling of min- 103971.
eral potential zones using data-driven based weighting factor and Forson, E.D., Wemegah, D.D., Hagan, G.B., Appiah, D., Addo-Wuver,
statistical index techniques. Journal of African Earth Sciences, 206, F., Adjovu, I., Otchere, F.O., Mateso, S., Menyeh, A. & Amponsah, T.
105020. (2022) Data-driven multi-index overlay gold prospectivity mapping
Amponsah, P.O., Forson, E.D., Sungzie, P.S. & Loh, Y. S.A. (2023) using geophysical and remote sensing datasets. Journal of African
Groundwater prospectivity modeling over the Akatsi districts in the Earth Sciences, 190, 104504.
Volta region of Ghana using the frequency ratio technique. Modeling Ghezelbash, R., Maghsoudi, A., Bigdeli, A. & Carranza, E.J.M.
Earth Systems and Environment, 9(2), 2081–2100. (2021) Regional-scale mineral prospectivity mapping: support vector
Amponsah, T.Y., Danuor, S.K., Wemegah, D.D. & Forson, E.D. (2022) machines and an improved data-drivenmulti-criteria decision-making
Groundwater potential characterisation over the Voltaian basin using technique. Natural Resources Research, 30(3), 1977–2005.
geophysical, geological, hydrological and topographical datasets. Griffis, R., Barning, K., Agezo, F. & Akosah, F. (2002) Gold deposits of
Journal of African Earth Sciences, 192, 104558. Ghana. Minerals Commission, Accra, Ghana, 432.
Annan-Yorke, R. & Cudjoe, J. (1971) Geology of the Voltaian basin GSD (2010) Geological map of Ghana. Geological Survey Department.
(summary of current ideas). Special Bulletin for Oil Exploration. Holden, E.-J., Wong, J.C., Kovesi, P., Wedge, D., Dentith, M. & Bagas,
Geological Survey Department, Accra, Ghana, 29. L. (2012) Identifying structural complexity in aeromagnetic data: an
Apambire, W.B. (2000) Geochemical modeling and geomedical implica- image analysis approach to greenfields gold exploration. Ore Geology
tions of fluoriferous groundwaters in the upper east region of Ghana. Reviews, 46, 47–59.
University of Nevada, Reno. Jenks, G.F. (1963) Generalization in statistical mapping. Annals of the
Arab Amiri, M., Karimi, M. & Alimohammadi Sarab, A. (2015) Hydro- Association of American Geographers, 53(1), 15–26.
carbon resources potential mapping using evidential belief functions Jordan, C., Carney, J., Thomas, C. & McDonnell, P. (2009) Ghana air-
and frequency ratio approaches, southeastern Saskatchewan, Canada. borne geophysics project in the Volta and Keta basins: BGS final
Canadian Journal of Earth Sciences, 52(3), 182–195. report. BGS commissioned report, CR/09/02. British Geological
Blakely, R.J., Cox, A. & Iufer, E.J. (1973) Vector magnetic data for Survey.
detecting short polarity intervals in marine magnetic profiles. Journal Kheyrollahi, H., Alinia, F. & Ghods, A. (2016) Regional magnetic
of Geophysical Research, 78(29), 6977–6983. lithologies and structures as controls on porphyry copper deposits:
Carney, J.N., Jordan, C.J., Thomas, C.W., Condon, D.J., Kemp, S.J. & evidence from Iran. Exploration Geophysics, 49(1), 98–110.
Duodo, J.A. (2010) Lithostratigraphy, sedimentation and evolution of Kinnear, J.A., Binley, A., Duque, C. & Engesgaard, P.K. (2013) Using
the Volta basin in Ghana. Precambrian Research, 183(4), 701–724. geophysics to map areas of potential groundwater discharge into
Carrier, M., Lefebvre, R., Racicot, J. & Asare, E. (2008) Northern Ringkøbing Fjord, Denmark. The Leading Edge, 32(7), 792–796.
Ghana hydrogeological assessment project. In Access to sanitation Kortatsi, B. (1994) Groundwater utilization in Ghana. IAHS
and safe water—Global partnerships and local actions: Proceed- Publications-Series of Proceedings and Reports-International
ings of the 33rd WEDC International Conference, Accra, Ghana. Association of Hydrological Sciences, 222, 149–156.
Loughborough, UK:WEDC, Loughborough University, pp. 353–361. Loh, Y.S.A., Akurugu, B.A.,Manu, E.&Aliou, A.-S. (2020) Assessment
Chegbeleh, L.P., Akudago, J.A., Nishigaki, M. & Edusei, S.N.K. (2009) of groundwater quality and the main controls on its hydrochemistry in
Electromagnetic geophysical survey for groundwater exploration in some Voltaian and basement aquifers, northern Ghana. Groundwater
the Voltaian of northern Ghana. Journal of Environmental Hydrology, for Sustainable Development, 10, 100296.
17, 1–16. MacDonald, A., Bonsor, H., Calow, R., Taylor, R., Lapworth, D.,
Darko, P.K. & Krásny, J. (2007) Regional transmissivity distribution and Maurice, L., Tucker, J. & O Dochartaigh, B. (2011) Groundwater
groundwater potential in hard rock of Ghana. In Krásný, J. & Sharp, J. resilience to climate change in Africa. British Geological Survey.
M. (Eds) Groundwater in fractured rocks. IAH Selected Paper Series, Mainoo, P.A., Manu, E., Yidana, S.M., Agyekum, W.A., Stigter, T.,
volume 9. London: CRC Press, pp. 125–136. Duah, A.A. & Preko, K. (2019) Application of 2D-electrical resistiv-
Dzigbodi-Adjimah, K. (1993) Geology and geochemical patterns of the ity tomography in delineating groundwater potential zones: case study
Birimian gold deposits, Ghana, West Africa. Journal of Geochemical from the Voltaian supergroup of Ghana. Journal of African Earth
Exploration, 47(1-3), 305–320. Sciences, 160, 103618.
 13652478, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1365-2478.13422 by University of Ghana - Accra, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
DEPTH-BASED CORRELATION ANALYSIS 15
Manu, E., Agyekum, W.A., Duah, A.A., Tagoe, R. & Preko, K. (2019) Roest,W.R., Verhoef, J. & Pilkington,M. (1992)Magnetic interpretation
Application of vertical electrical sounding for groundwater explo- using the 3-D analytic signal. Geophysics, 57(1), 116–125.
ration of Cape Coast municipality in the Central Region of Ghana. Sabins, F.F. (1999) Remote sensing for mineral exploration.Ore Geology
Arabian Journal of Geosciences, 12(6), 1–11. Reviews, 14(3-4), 157–183.
Martin, N. & Van De Giesen, N. (2005) Spatial distribution of ground- Sankar, K. (2002) Evaluation of groundwater potential zones using
water production and development potential in the Volta River basin remote sensing data in Upper Vaigai river basin, Tamil Nadu, India.
of Ghana and Burkina Faso. Water International, 30(2), 239–249. Journal of the Indian Society of Remote Sensing, 30(3), 119–129.
Mul, M., Obuobie, E., Appoh, R., Kankam-Yeboah, K., Bekoe-Obeng, Service, G.S. (2014) 2010 population and housing census report. Accra:
E., Amisigo, B., Logah, F.Y., Ghansah, B. & McCartney, M. (2015) Ghana Statistical Service.
Water resources assessment of the Volta River basin. IWMI Work- Smith, R.S., O’Connell, M.D. & Poulsen, L.H. (2004) Using airborne
ing Paper 166. Colombo, Sri Lanka: InternationalWater Management electromagnetics surveys to investigate the hydrogeology of an area
Institute. near Nyborg, Denmark. Near Surface Geophysics, 2(3), 123–130.
Mulwa, J., Fairhead, D., Barongo, J. & Mariita, N. (2009) Heat Srivastava, P.K. & Bhattacharya, A.K. (2006) Groundwater assessment
source mapping and evaluation of geothermal resource potential in through an integrated approach using remote sensing, GIS and resis-
Lake Bogoria Basin, Kenya. In: SEG Technical program expanded tivity techniques: a case study from a hard rock terrain. International
abstracts 2009. Houston, TX: Society of Exploration Geophysicists, Journal of Remote Sensing, 27(20), 4599–4620.
pp. 1294–1299. Vakhshoori, V. & Zare, M. (2016) Landslide susceptibility mapping
Nabighian, M.N., Ander, M., Grauch, V., Hansen, R., LaFehr, T., Li, by comparing weight of evidence, fuzzy logic, and frequency ratio
Y., Pearson, W., Peirce, J., Phillips, J. & Ruder, M. (2005) Historical methods. Geomatics, Natural Hazards and Risk, 7(5), 1731–1752.
development of the gravity method in exploration. Geophysics, 70(6), Yidana, S.M. (2010) Groundwater classification using multivariate sta-
63ND–89ND. tistical methods: Southern Ghana. Journal of African Earth Sciences,
Nsiah, E., Appiah-Adjei, E.K. & Adjei, K.A. (2018) Hydrogeologi- 57(5), 455–469.
cal delineation of groundwater potential zones in the Nabogo basin, Yidana, S.M., Vakpo, E.K., Sakyi, P.A., Chegbeleh, L.P. & Akabzaa,
Ghana. Journal of African Earth Sciences, 143, 1–9. T.M. (2019) Groundwater–lakewater interactions: an evaluation of the
Nwachukwu, C., Orakwe, L. & Ogbu, K. (2018) Review of soil and impacts of climate change and increased abstractions on groundwater
water assessment tool (SWAT) application in sub-Saharan African contribution to the Volta Lake, Ghana.Environmental Earth Sciences,
countries. Journal of Engineering and Applied Sciences, 13, 1–10. 78(3), 74.
Obuobie, E., Barry, B. & Agyekum, W. (2016) Groundwater resources
of the Volta basin. In: Williams, T.O., Mul, M., Biney, C.A., &
Smakhtin, V. (Eds) The Volta River basin: Water for food, economic
growth and environment. London: Routledge, pp. 66–81. How to cite this article: Yaw Amponsah, T.,
Preeja, K., Joseph, S., Thomas, J. & Vijith, H. (2011) Identification of Wemegah, D.D., Danuor, S.K. & Forson, E.D. (2023)
groundwater potential zones of a tropical river basin (Kerala, India) Depth-based correlation analysis between the density
using remote sensing and GIS techniques. Journal of the Indian of lineaments in the crystalline basement’s weathered
Society of Remote Sensing, 39(1), 83–94.
zones and groundwater occurrences within the
Reeves, C. (1989) Aeromagnetic interpretation and rock magnetism.
First Break Voltaian basin, Ghana. Geophysical Prospecting,, 7(7).
Roberts, N. (1982) A note on the geomorphological environment of Çatal 1–15. https://doi.org/10.1111/1365-2478.13422
Hüyük, Turkey. Journal of Archaeological Science, 9(4), 341–348.
 13652478, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1365-2478.13422 by University of Ghana - Accra, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License