Sustainable Water Resources Management (2022) 8:55

https://doi.org/10.1007/s40899-022-00639-8

ORIGINAL ARTICLE

Delineation of groundwater potential zones using cokriging

and weighted overlay techniques in the Assin Municipalities of Ghana
Dennis Asante1 · Emmanuel K. Appiah‑Adjei1 · Albert Asare1

Received: 2 August 2021 / Accepted: 9 February 2022 / Published online: 3 March 2022
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2022

Abstract
This study has employed cokriging and weighted overlay techniques in a GIS environment to delineate groundwater potential
zones from hydrogeological parameters in the Assin Municipalities, Ghana, where groundwater is the main source of pota-
ble water but access to it is less predictable since the aquifers are structurally controlled. Data from drilling logs, borehole
development and construction, and pumping test on 67 boreholes in the area were utilized to estimate the hydrogeological
parameters such as borehole yield, transmissivity, specific capacity, aquifer thickness and water strike elevation. Then, using
the cokriging and weighted overlay techniques in ArcGIS, spatial thematic maps were created for all the parameters and
integrated to produce the groundwater potential map, which categorized the study area into very low, low, moderate, and
high groundwater potential zones. The results revealed that the cokriging technique was better in delineating the groundwater
potential zones of the study area with a prediction accuracy of 67%, while the weighted overlay approach had an accuracy of
44%. However, the delineated moderate and high groundwater potential zones in the area for both approaches were largely the
same and underlined by granitic rocks. Also, the generated groundwater potential map categorized approximately 0.1% (2.78
­km2), 35.2% (850.45 ­km2), 43.6% (1054.62 ­km2) and 21.1% (509.15 ­km2) of the study area as very low, low, moderate and
high groundwater potential zones, respectively. Thus, the study provides very useful information on groundwater potential
zones delineation, which would aid in effective exploration and development of groundwater resources.

Keywords Cokriging · Weighted overlay · Groundwater potential · Hydrogeological parameters · Assin Municipalities

Introduction drainage pattern, topography, slope and land use/land cover

(LULC) of the area. Thus, detailed investigations, ranging
Over 50% of global population depend on groundwater as from remote sensing, geological, geophysical, and hydrogeo-
the source of their drinking water (FAO 2002), whereas logical techniques, are required in identifying appropriate
about 2.5 billion people worldwide depend solely on ground- locations for sustainable abstraction of the resource in any
water resources for their daily water needs (UNESCO 2012) area. This is to avoid loss of investments associated with
due to its abundance, suitable natural quality, and availabil- drilling unsuccessful boreholes/wells, drying up of bore-
ity in all seasons. In Ghana, groundwater serves as the main holes/wells in periods of prolonged absence of rainfall and,
source of potable water for about 50% of the populace living sometimes, unacceptable water quality (Kortatsi 1994; Xu
in rural areas and it is increasingly becoming the only water and Usher 2006).
source available to developing communities. Although it is Hydrogeological studies such delineation of ground-
known to be the most abundant freshwater resource glob- water potential zones and groundwater modelling involve
ally, its availability for abstraction and sustainable use in an large volumes of multidisciplinary data (Gogu et al. 2001).
area depends on certain controlling factors including geo- Thus, Geographic Information Systems (GIS) has become
logic media, hydrogeological properties, climatic conditions, a common tool used in such studies due to its suitability for
handling complex geo-referenced data and applications in
* Dennis Asante several disciplines including engineering and environmental
[email protected] studies (Stafford 1991; Goodchild 1993). Several research-
ers (Waikar and Nilawar 2014; Preeja et al. 2011; Yeh
1
Geological Engineering Department, Kwame Nkrumah et al. 2009; Sener et al. 2005; Sikdar et al. 2004; Srinivasa
University of Science and Technology, Kumasi, Ghana

13
Vol.:(0123456789)
55 Page 2 of 15 Sustainable Water Resources Management (2022) 8:55

and Jugran 2003; Jaiswal et al. 2003; Shahid et al. 2000; which could be very subjective. Nonetheless, researchers
Krishnamurthy et al. 2000, etc.) have used remote sensing including Zare et al. (2014) and Yalcin (2005) have reported
and GIS techniques in delineating groundwater potential that cokriging generally improves the prediction accuracy of
zones in several parts of the world. Oh et al. (2011) used GIS spatial data interpolation. This is achieved by modelling the
and a probability model to map the probabilistic groundwa- variogram of each parameter, and then the cross-variogram
ter potential while Magesh et al. (2012) delineated ground- of the primary data/parameter and the covariate(s). Thus,
water potential zones using remote sensing, GIS and MIF cokriging makes use of secondary variables to improve upon
techniques. On the other hand, other researchers have used the prediction accuracy (Yalcin 2005). As a result, Giraldo
integrated remote sensing, GIS and geoelectrical techniques et al. (2020) used it to propose a model for predicting the
(vertical electrical sounding, VES) to delineate groundwater environmental pollution of particulate matter considering
potential zones where the VES was used to determine sub- wind speed curves as functional secondary variable. Hoosh-
surface parameters such as aquifer resistivity, aquifer thick- mand et al. (2011) also used cokriging in spatial estimation
ness, and depth to bedrocks (Chowdhury et al. 2009; Sriv- of groundwater quality parameters, whereas Ahmadi and
astava and Bhattacharya 2006; Israil et al. 2006; Sreedevi Sedghamiz (2008), on the other hand, applied cokriging to
et al. 2005; Rao 2003; Shahid and Nath 2002; Murthy 2000; map the groundwater depth. Similar works have been done
Edet and Okereke 1997). However, Todd (1980), as well as by Hoeksema et al. (1989) Deutsch and Journel (1992), and
Jha and Peiffer (2006), stated that the existence of ground- Goovaerts (1997) using cokriging.
water in an area can only be inferred from surface features This study, therefore, delineates the groundwater poten-
such as geology, geomorphology, soils, land use/land cover, tial zones in the Assin municipalities by integrating various
surface water bodies, drainage pattern, topographic slope, hydrogeological parameters using both cokriging and the
etc., which act as indicators of groundwater existence in weighted overlay approahes in ArcGIS environment. The
remotely sensed data from satellite imagery since remote accuracy of groundwater potential maps generated by the
sensors cannot directly detect the presence of groundwater. two approaches are compared to determine the method that
As such, test drilling and stratigraphy analyses remain the better predicts the groundwater potential of the area. The
most reliable and standard methods used for determining the study area has over 85% of the population residing in rural
location of aquifer, groundwater quality, physical character- areas with boreholes and protected hand-dug wells con-
istics of aquifers, etc. in an area (Chowdhury et al. 2009), tributing over 90% of domestic water supplies (GSS 2012).
although they are costly, time-consuming and require skilled Geologically, it is predominantly underlined by crystalline
manpower (Sander et al. 1996). basin-type granitoids (GSA 2009) with a dense network of
The use of GIS techniques in groundwater potential fractures (Ewusi and Kuma 2011); hence, the occurrence of
assessment varies considerably from one researcher to groundwater in the area is less predictable. Thus, delineating
another. Whereas some researchers use the local influence the groundwater potential zones of the area is envisaged to
of groundwater controlling parameters in assigning weights provide useful prior information to aid in effective explora-
to different thematic layers prior to their integration for gen- tion and exploitation of the resource. Additionally, it would
eration of groundwater potential map of an area (Nsiah et al highlight areas where groundwater developers need to carry
2018; Gumma and Pavelic 2013), others use the Analytical out more comprehensive groundwater exploration to avoid
Hierarchy Process (AHP) of the GIS-based multi-criteria drilling unsuccessful boreholes and, thereby, safeguard their
decision analysis (MCDA) for the normalization of the investments.
assigned weights (Yifru et al. 2020; Arulbalaji et al. 2019;
Andualem and Demeke 2019; Mallick et al. 2019; Bashe
2017; Hussein et al. 2017; Pinto et al. 2017; Shahid and Nath Materials and methods
2002). Also, the weighted overlay approach in groundwater
potential delineation involves reclassification and weighting Study area
of the input thematic layers of the hydrogeological param-
eters prior to their integration. However, not all such param- The study area, located in the Central Region of Ghana
eters (e.g. aquifer zone thickness, fractured zone thickness, (Fig. 1), consists of the Assin South, Assin Central and Assin
static water level, overburden thickness, etc.) have a stand- North Municipalities, and lies within longitudes 5° 20ʹ N to
ard classification scheme for such purpose. Hence, different 6° 00ʹ N and latitudes 1° 06ʹ W to 1 40ʹ W, covering a total
researchers use different ranges of classification for the same land area of about 2417 ­km2. The main source of water for
parameters in different scenarios based on the local geol- domestic purposes are boreholes fitted with hand pumps and
ogy and some other hydrogeological consideration (Anteneh protected Hand-Dug Wells (HDW), with pipe borne water
et al. 2021; Kumar et al. 2019; Nsiah et al. 2018; Akinlalu contributing less than 10% (GSS 2012). Topography of the
et al. 2017; Srinivasa and Jugran 2003; Jhariya et al. 2016), area is generally rolling to undulating with slopes trending

13
Sustainable Water Resources Management (2022) 8:55 Page 3 of 15 55

Fig. 1  Study area location with underlying geology and distribution of study boreholes

southwards from the Birim River, which is the main drain- network of fractures are commonly found near the surface
age system in the area and a major tributary of the Pra River rather than at greater depths. However, the degree of weath-
(Asante-Annor et al. 2018). It falls within the evergreen and ering depends on the mineralogical composition of rocks,
semi-deciduous forest zones with thick virgin forest in areas degree of fracturing and the rate of precipitation.
where there are forest reserves and lies within the semi- In general, the targets zones for groundwater develop-
equatorial climate (Dickson et al. 1988), which is character- ment projects in the study area are weathered zones, rocks
ized by bimodal rainfall pattern (Asante-Annor et al. 2018). greatly affected by tectonic activity, contacts of intrusions,
Predominantly, the study area is made up of the Basin and highly fractured silica-rich and silicified rocks (Asante-
type granitoid of the Birimian System, covering about 90% Annor et al. 2018). However, the most productive area for
of the area. The rock units found in the area include gneiss, groundwater is a formation with a combination of thick
granites and granodiorites with the gneissic rocks intruded weathered zone and well-formed and highly persistent frac-
by both acidic and basic igneous rocks such as white and ture network in the bedrock (Prackley 1984).
pink pegmatite, aplite, granodiorite, and dolerite dykes
(Gibrilla et al. 2010; Ganyaglo et al. 2010). The north-west- Data acquisition and database creation
ern portion of the area is made up of the Banket group of the
Tarkwaian System consisting of rocks such as conglomerate, Data on seventy-seven (77) boreholes drilled in the study
quartz pebble and quartzose sandstones (GSD 2009). Rocks area between the years 2006 and 2008 were acquired for the
in the study area have negligible primary porosity and per- study from the Community Water and Sanitation Agency
meability due to their crystalline nature; thus, groundwater (CWSA) in Ghana. Pertinent information contained in the
flow is controlled by secondary geological structures such data included borehole location coordinates, borehole state
as fractures, joints and shear zones, and weathering in the (successful or unsuccessful), diameter and depth of bore-
rocks. Ewusi and Kuma (2011) indicated that weathering hole, geological drilling logs (from which parameters such
of the granitoids in the area is relatively thin, and dense as depth to bedrock, aquifer zone thicknesses, fractured zone

13
55 Page 4 of 15 Sustainable Water Resources Management (2022) 8:55

thickness, water strike, geology and lithologic description the interpolation method used for this study requires that the
were determined), borehole development and construction data be normally distributed. However, if the data distribu-
with airlift yield information, pumping test data with static tion is otherwise, there are inbuilt mathematical functions
water levels, and water quality tests information. Overall, to transform the data. Also, there should not be trends in
sixty-seven (67) of the drilled boreholes were successful the data to produce a more accurate interpolation surface.
(i.e., yields greater 13.5 L/min per CWSA guidelines), while Hence, the histogram and Normal Quantile–Quantile (QQ)
the ten (10) were marginal to dry holes. The Mean Sea Level plot tools in ArcGIS were used together with the results from
(MSL) was used as a common reference point in assess- the data distribution analyses from the SPSS (skewness and
ing the general variation of parameters such as static water kurtosis) to study the distribution of each hydrogeological
level (SWL), water strike and depth to bedrock by subtract- parameter, whilst the trend analysis tool was used to deter-
ing them from their respective ground elevations to obtain mine whether trends exist in the data. Parameters that were
the Static Water Elevation (SWE), Water Strike Elevation not normally distributed were transformed using the appro-
(WSE) and Bedrock Elevation, respectively. priate mathematical module whereas those parameters that
The aquifer parameters like transmissivity, specific showed trends were corrected by removing them using the
capacity, and maximum potential (sustainable) borehole appropriate trend removal module.
yield were computed from the pumping test data. Each of Ordinary kriging interpolation technique was used to gen-
the pumping tests was conducted over a 9 h duration, com- erate the spatial distribution layers for the individual hydro-
prising 6 h of pumping at a constant rate followed by 3 h geological parameters. Generally, geologic field data such
of recovery, in conformity with CWSA guidelines (CWSA as those used for this work are usually anisotropic in nature.
2010) and they were all single well tests. Due to well loss However, several researchers (Eldeiry and Garcia 2012; Hu
problems that is commonly associated with drawdown meas- et al. 2005) have observed the usefulness of ordinary kriging
urements in single well tests, data from the recovery phase of interpolation technique in environmental interpolation map-
the pumping tests were used to estimate the Transmissivity ping, groundwater quality assessment studies and soil sci-
(T) values following the Theis’ (1935) recovery approach. ence due to its ability to account for anisotropy in the data-
Finally, the maximum potential (sustainable) yield of each set. Therefore, for the purpose of validating the above claim
borehole was computed using Eq. (1) proposed by van and for comparison, other kriging methods such as simple
Tonder et al. (2000). and universal kriging were also used to produce prediction
maps of each parameter. It was, however, observed that the
QSustainable = (QPumping ∕SPumping ) × SAvailable , (1) ordinary kriging produced a better interpolation surface.
where the ratio QPumping ∕SPumping is the specific capacity of The dataset for each parameter was randomly split into
the aquifer, with QPumping being the discharge rate Q, and two (2) parts, each consisting of 70% training data, which
SPumping being the total drawdown at the end of the pumping. was used in the creation of the prediction maps, and 30%
The available drawdown ­(sAvailable) is given by Eq. (2): test data used to validate the maps generated. The spatial
autocorrelation of each dataset was predicted using the
SAvailable = Pump setting − SWL − buffer. (2) empirical semivariogram model of the geostatistical tool
in ArcGIS software. In kriging, the model type used influ-
A column of water called the buffer is always left above ences the output prediction surface as each model has been
the pump setting beyond which the drawdown should not designed to fit different type of phenomena more accurately
exceed. However, not all the borehole logs had the main (ESRI 2016). Thus, the transformation and model type used
water strike clearly indicated on them. Thus, for the sake for each parameter was based on the transformation—model
of consistency and to ensure sustainable pumping of each type combination that produced the best prediction surface
borehole, a buffer of 3 m was used throughout in the deter- as per the results from validation and cross-validation. The
mination of the sustainable borehole yields. performance of each model was assessed using both valida-
tion and cross-validation. In validation, the test data of a
Generation of thematic maps of the hydrogeological parameter was plotted on the geostatistical interpolation sur-
parameters face produced by each model. The predicted values for those
points from the interpolated surface were then extracted
Figure 2 shows the sequential order followed in the gen- using the “Extract to Points” tool in ArcGIS. The extracted
eration of thematic maps for hydrogeological parameters predicted values were then compared with the measured test
used in this study. Descriptive statistics such as the mean, data values by determining the Pearson correlation coeffi-
mode, standard deviation, range, skewness, and kurtosis cient (r) between them. Also, the accuracy of the generated
were computed for each hydrogeological parameter using surface for each parameter was determined by expressing the
SPSS version 20. The mathematical principles underlying number of test data that fell within their exact zones on the

13
Sustainable Water Resources Management (2022) 8:55 Page 5 of 15 55

Data Collection / Database Creation

Determination of Hydrogeological Parameters

Spatial Statistics

100 % Dataset Randomly Split 100 % Dataset

Interpolation of Surficial maps 70 % Training 30 % Test Data

by Kriging Data

70 % Accuracy Validation
Surficial Maps
Matrix from
Cross - Validation
Accuracy Matrix
Cokriging and
Weighted Overlay
Integration
Compare to 70%
Test Data
Non – Accuracy Matrix
Spatial Groundwater
Exploratory Potential Maps
Statistics

Fig. 2  The methodology followed in the generation of maps in the study area

generated prediction surface as percentage of the total test then classified into their appropriate classes. The classifica-
data. In the case of cross-validation, the mean standardized tion of the values of the input raster layers were based on
Error (ME), root mean square standardized error (RMSE), specified scheme. For instance, categorization scheme for
root mean square (RMS), and average standard error (ASE) classifying the transmissivity raster layer was according to
were used to assess the performance of each model. the Krasny’s classification scheme (Krasny 1993), whereas
The groundwater potential map of the area was created that of the yield layer was according to the CWSA (2010)
using both Cokriging and the Weighted Overlay approaches guidelines. The classified feature classes were then reclassi-
to determine which method better predicted the groundwa- fied by ranking them on a scale of 1–5 based on the param-
ter potential of the study area. It has been established by eter’s influence/indication of the availability of groundwater
number of researchers including Mohammed-Aslam et al. or otherwise, with 5 indicating the highest possibility of the
(2010) that there exists a positive correlation between bore- presence of groundwater, whilst 1 indicates the least. Appro-
hole yield and availability of groundwater in any given area; priate weights were then assigned to the reclassified raster
hence, the Spearman’s Correlation Coefficient (ρ) between layers based on their relative influence on the occurrence
borehole yield and each individual parameter was computed. of groundwater and in accordance with similar research by
The yield layer was used as primary variable of interest and Nsiah et al. (2018), and further integrated in ArcGIS to gen-
integrated with different combinations of the other param- erate the composite groundwater potential maps using the
eters that correlates well with the yield layer as covariates various parametric combinations. For the sake of consist-
to produce the groundwater potential map of the study area ency, each of the parametric combinations used in cokriging
using cokriging in ArcGIS. was repeated in the weighted overlay approach. Additional
In the weighted overlay approach, the interpolated the- potential maps were generated using the weighted overlay
matic layers of the aquifer parameters were extracted into approach by adding more parameters to that used already to
raster format in ArcGIS. The extracted raster layers were

13
55 Page 6 of 15 Sustainable Water Resources Management (2022) 8:55

determine its effect on the prediction accuracy of the gener- Table 1  Parametric combinations and the weights assigned to each
ated potential maps. layer
Also, the weights assigned to the individual layers used in Map number Parametric combination Weight (%)
each of the models for the weighted overlay potential maps
Map 1 Yield 40
were varied until the ones that produced the best potential
Specific capacity 25
surface for each parametric combination as per the validation
Transmissivity 25
results were obtained. Table 1 shows the parametric combi-
Static water level 10
nations and the weights assigned to each reclassified raster
Map 2 Yield 40
layer in the generation of groundwater potential maps using
Specific capacity 30
the weighted overlay approach. The groundwater potential
Water strike 20
maps were validated by plotting the yield test dataset on each
Aquifer zone thickness 10
generated layer and expressing the number of data points
Map 3 Yield 40
that fell within the right zones delineated as low, moder-
Specific capacity 25
ate, high, and very high zones as a percentage of the total
Transmissivity 20
number of data points. Classification of yield values as low,
Aquifer zone thickness 15
moderate, high, etc. was in accordance with the guidelines
Map 4 Yield 35
provided by the Community Water and Sanitation Agency in
Specific capacity 20
their sector guidelines for small communities (CWSA 2010).
Transmissivity 20
Aquifer zone Thickness 15
Water strike 5
Results and discussion
Total drawdown 5
Map 5 Yield 30
Exploratory statistics of the aquifer parameters
Specific capacity 20
Transmissivity 15
Table 2 shows results for the non-spatial exploratory analy-
Aquifer zone Thickness 10
ses of the hydrogeological data. Correlation between the
Static water elevation 10
ground elevation and bedrock elevation using the Spear-
Overburden thickness 5
man’s Coefficient of correlation (ρ) shows a strong posi-
Total drawdown 5
tive correlation (0.972), indicating high bedrock elevation
Percentage recovery 5
in areas with high ground elevations and vice versa. This
is confirmed by a weak negative correlation (ρ = − 0.291)
between overburden thickness and bedrock elevation, and
thus, weathering effects on bedrock elevation in the area is the boreholes have recovery percentages greater than the
minimal. Also, the borehole depth is not dependent on the mean value of 85%; hence, good for domestic water supply.
depth of overburden. Though the thickness of overburden Results from the exploratory spatial statistics (Table 3)
did not exceed 35 m, depth to fresh rock ranged from 12 to indicate that apart from specific capacity dataset, which
55 m, having an average of about 35 m. Thus, on the aver- had strong positive correlation (ρ = 0.866) with the yield
age, about 20–25 m of the bedrock has undergone some level dataset, all the other parameters showed weak to no correla-
of alteration ranging from slight to moderate weathering. A tion with the yield dataset. Nonetheless, transmissivity and
very weak positive correlation (ρ = 0.192) exists between the SWL showed relatively better correlation with yield than
fractured zone and the aquifer zone thicknesses contrary to the other parameters. Arguably, WSE, SWL and total draw-
water strike elevation and static water level, which have a down datasets showed negative correlation with the yield
strong positive correlation (ρ = 0.926). It was observed that dataset as expected. The cross-validation matrices (Table 3)
the values of SWE are generally higher than their respective from the various models produced ME values, which cen-
WSE, suggesting that the aquifers in the area are semi-con- tred around zero (0) for all the parameters with the excep-
fined to confined (Buckley 1986; Gibrilla et al. 2010). Both tion of SWL that produced a relatively higher ME value. In
transmissivity and specific capacity datasets were right- essence, negative values of ME for borehole depth, over-
skewed, with a moderate positive correlation (ρ = 0.572) burden thickness, water strike, SWL, specific capacity, and
between them. Borehole sustainable yield was also posi- transmissivity models indicate underestimation of the errors
tively skewed and recorded the highest skewness among in the prediction surface produced by their respective mod-
all the parameters. However, the borehole recovery dataset els, whereas positive ME values for all the remaining models
showed negative skewness (− 0.66), indicating that most of are indication of overestimation of the errors. In the case of
RMSE, almost all the models produced values closer to the

13
Sustainable Water Resources Management (2022) 8:55 Page 7 of 15 55

Table 2  Summary statistics for hydrogeological parameters

Parameter Count Mean Median Range Standard Variance Skewness Kurtosis No. of
deviation Outli-
ers

Ground elevation (m) 77 148.48 149 57–200 26.95 726.76 − 0.37 0.694 0
Borehole depth (m) 77 52.19 45 15–80 15.14 229.15 0.24 − 0.92 0
Overburden thickness (m) 77 13.38 13 3–35 6.64 44.03 0.92 1.05 9
Bedrock elevation (m) 77 135.04 136 43–194 25.18 792.88 -0.33 0.76 1
Depth to fresh rock (m) 77 34.57 36 12–55 7.37 54.35 − 0.43 1.35 3
Fractured zone thickness (m) 77 21.99 23 4–36 7.29 53.15 − 0.08 − 0.58 0
Aquifer zone thickness (m) 55 19.96 19 4–43 9.24 85.37 0.59 − 0.31 0
Water strike (m) 59 30.63 28 12–62 10.4 108.10 1.34 2.07 4
Water strike elevation (m) 59 118.3 119.4 32.3–186.1 29.59 875.52 − 0.2 0.33 1
Static water level (m) 36 7.10 6.12 1.1–25.60 5.39 29.05 1.64 3.09 3
Static water elevation (m) 36 144.1 144.5 84.7–197.0 28.57 816.03 0.12 -0.81 0
Total drawdown (m) 36 14.35 13.95 4.0–30.5 5.81 33.79 0.64 0.45 1
Percentage recovery (%) 36 86.26 89.73 58–100 13.72 188.37 − 0.66 − 0.86 0
Sustainable yield (L/min) 36 92.05 63.82 8.5–427.4 92.03 8469.05 1.84 3.86 2
Specific capacity ­(m2/day) 36 4.98 3.49 0.3–16.9 4.54 20.65 1.27 0.66 2
Transmissivity ­(m2/day) 36 6.27 3.15 0.1–28.2 7.19 51.77 1.72 2.54 2

Table 3  Accuracy matrices and Parameter ME ASE RMS RMSE r Accuracy (%)
validation results from kriging
of each parameter Borehole depth − 0.001 12.673 13.587 1.067 0.133 36.84
Overburden thickness − 0.027 7.526 6.295 0.815 0.000 61.11
Bedrock elevation 0.026 25.325 27.631 1.088 0.771 52.63
Depth to fresh rock 0.056 7.589 7.416 0.967 0.218 57.89
Fractured zone thickness 0.003 7.922 8.010 1.011 0.142 30.00
Water strike − 0.040 10.444 10.791 0.992 0.359 50.00
Aquifer zone thickness 0.003 9.640 9.413 0.977 0.434 42.86
Water strike elevation 0.015 25.979 28.255 1.050 0.413 57.14
Static water level − 0.154 6.798 6.686 1.101 − 0.677 37.50
Static water elevation 0.046 31.335 33.063 1.050 0.262 37.50
Total drawdown 0.058 6.450 6.893 1.063 0.229 50.00
Percentage recovery 0.013 14.506 13.567 0.916 0.772 62.50
Sustainable yield 0.009 154.70 81.477 0.546 0.672 62.50
Specific capacity − 0.062 8.408 4.713 -0.706 0.197 77.78
Transmissivity − 0.010 58.268 10.556 0.292 − 0.076 66.67

expected value of one (1) with only the sustainable yield and underestimated the variability of the predicted values from
transmissivity models recording values, which moderately the measured values. Contrary to this, the models for over-
differed from one (1); thus, making the model prediction burden thickness, depth to bedrock, aquifer zone thickness,
errors invalid. Models for all parameters recorded negleg- SWL, percentage recovery, sustainable yield, specific capac-
ible difference between RMS and ASE (which is ideal for ity and transmissivity have ASE > RMS, indicating overes-
a well-fitted model) with the exception of sustainable yield timation of the variability of the predicted values from the
and transmissivity models, which recorded considerable dif- measured values. Bedrock elevation, percentage recovery
ference between their respective estimated RMS and ASE. and sustainable yield recorded a strong correlation between
Notwithstanding this, ASE < RMS for models of borehole the predicted and measured test data values with percentage
depth, bedrock elevation, fractured zone thickness, water recovery recording the highest. Water strike, aquifer zone
strike, WSE, SWE and total drawdown implies the models thickness and WSE recorded weak correlation between

13
55 Page 8 of 15 Sustainable Water Resources Management (2022) 8:55

Fig. 3  Overlay of a borehole depth and b overburden thickness on the study area geology

the predicted and measured test data values, whereas all The deepest boreholes are in areas underlained by the undif-
the remaining parameters recorded weak to no correlation ferentiated granitoids and the biotite gneiss.
between their respective measured and predicted values. Overburden thickness within the area is generally between
There was no correlation between the measured and pre- 9 and 18 m with isolated patches of areas above 18 m and
dicted test data values for overburden thickness whilst SWL below 9 m, respectively, located around the northern portion
recorded a negative correlation. underlained by the biotite gneiss and south-eastern part of
From the validation results, however, the specific capacity the area (Fig. 3b). This range of values is consistent with
model had the best accuracy of 78% as opposed to the 63% that reported by Gibrilla et al. (2010) of between 4 and 20 m
of percentage recovery, which otherwise recorded the high- within the basin granitoids, which further attributes the wide
est correlation coefficient between the predicted and meas- range in overburden thickness to varrying climatic condi-
ured test data values. Although the overburden thickness tions in the area. Gibrilla et al. (2011) noted that borehole
and transmissivity models each had no correlation between yield in the intrusive granite complexes (granitoids) is gen-
their measured and predicted test data values, each recorded erally low in areas where depth of weathering is shallow.
an accuracy of over 60% from the validation. Also, bedrock Ewusi and Kuma (2011) and Ganyaglo et al. (2012) indi-
elevation, aquifer zone thickness, percentage recovery and cated that the depth of weathering is generally thicker in
sustainable yield showed strong agreement between the cor- sedimentary and volcanic terranes as compared to granitoid
relation and validation results whereas water strike, WSE terranes, which is not entirely the case in the current study.
and SWE showed a moderate agreement. The depth of weathering in an area is, however, subject to
climatic condition, particularly amount, distribution and fre-
Spatial evaluation of hydrogeological parameters quency of rainfall, the extent and distribution of fractures
and other structural discontinuities in the area (NRC 1996).
The depth range of boreholes (Fig. 3a) obtained in this study Varying climatic conditions in the area (Gibrilla et al. 2011)
range from 15 to 80 m, which is consistent with ranges pro- and varying degree of structural discontinuities may, there-
vided in other studies such as 35–90 m by Asante-Annor fore, have caused the discrepancy above.
et al. (2018), 23–40 m by WRI (1992), and 15–100 by From Fig. 4a, it is observed that the north-western part
Armah (2000). Three distinct depth zones can be observed of the area, underlained by the Banket group, recorded the
on the spatial variation plot of borehole depth, viz. ≤ 50 m, lowest bedrock elevations. The granitoid terrains generally
50–60 m, and > 60 m. It is observed that the range of bore- have bedrock elevations between 130 and 150 m above
hole depth in the northern part of the area is generally below MSL, whereas the Birimian metasedimentary terrains
55 m. The zone with the shallowest depths is underlained by (the sericite schist and biotite gneiss zones) have bedrock
leucogranite and portions of the undifferentiated granitoids, elevation of about 120–140 m above MSL. Three (3) dis-
which are parts of the granitoids of the Birimian supergroup. tinct fractured zone thickness ranges can be observed in
the study area, viz. ≤ 16.5 m, 16.5–23 m, > 23 m (Fig. 5a).

13
Sustainable Water Resources Management (2022) 8:55 Page 9 of 15 55

Fig. 4  Overlay of spatial distribution of a bedrock elevation, b fractured zone thickness thickness on the study area geology

Fig. 5  Overlay of spatial distribution of a water strike depth and b water strike elevation on study area geology

Generally, all the thickness ranges can be found in the drilling. Aquifers in crystalline rocks are mainly fractured
undifferentiated granite and the biotite gneiss terrains, zone aquifers and are developed at depth of about 20 m or
whereas the Banket formation and the leucogranite zones more below ground surface (Gibrilla et al. 2011). Thus, the
have ranges of about 20–25 m and 16.5–23 m, respectively. depth range of aquifers encountered in the area is consistent
Depth at which water was struck in boreholes is generally with values provided by Asante-Annor et al. (2018). Fig-
high (generally > 30 m) at the south-eastern part of the area ure 5b is the elevation of the aquifers encountered in the
underlained by the biotite gneiss and undifferentiated bio- area. It is observed that the deepest aquifers are at relatively
tite granite, and gradually decrease towards the western and lower elevations of 78–107 m above MSL. Gibrilla et al.
north-eastern portions. Water strike in the undifferentiated (2011) stated that fractured zone aquifers in crystalline rocks
granites, which covers the greater part of the area, is gener- tend to be localized in nature and, hence, groundwater occur-
ally around 33 m or below whereas that of the leucogranite rences are controlled by degree of fracturing and nature
are below 28 m (Fig. 5a). As indicated on the borehole logs, of groundwater recharge, whereas Ganyaglo et al. (2012)
water was encountered mainly in fractured zones during argued that fractured zone aquifers may be continuous or

13
55 Page 10 of 15 Sustainable Water Resources Management (2022) 8:55

Fig. 6  Overlay of spatial distribution of a SWL and b SWE on the study area geology

discontinuous depending on the nature of the prevailing geo- that most of the borehole locations of the work of Asante-
logical structures. Annor et al. (2018) are within the granitod zones delineated
Generally, SWL in the area is less than 10 m, with a small as high yield zones in the current research; hence, it could
localized area within the north-western corner underlain by possibly be the reason for the apparent deviation in the lower
sericite schist and the Banket formations having SWL values limit of the range.
greater than 10 m (Fig. 6a). This range of values of SWL Prackley (1984) generally reported the borehole yields
are consistent with the range of 1.09–14.75 m reported by within the entire Cape Coast basin as 1.67–500 L/min
Asante-Annor et al (2018). It was observed in this study that with a mean of 33.33 L/min for the intrusive granites, and
across a given geological formation in the area, the SWL 11.66–150 L/min with a mean of 61.67 L/min for the Birim-
varies over just a small range. Figure 6b shows the spatial ian. Geologically, the high yielding boreholes are located
distribution of SWE in the area. Similar to the SWL, SWE within the undifferentiated granite zones and parts of the
does not show much variation over a given geological for- biotite granite zones. The Banket formation, the sericite
mation. Also, the pattern of distribution of SWE is similar schist, the north-western part of the undifferentiated granites
to that of SWL since most of the areas with shallow SWL and the northern part of the leucogranite are the terrains that
have relatively higher SWE. However, the spatial variation recorded low yields.
of the SWE is a bit more uniform compared to the SWL. The Transmissivity values were categorized using the Kras-
highest SWE values occurred around the north-eastern part ny’s classification scheme (Krasny 1993) and the spatial
of the area underlained by the leucogranites. Hence, given distribution is shown in Fig. 7c. Generally, the transmis-
that there is a hydraulic connection, groudwater is expected sivity ranges from 0.1 to 28.2 m ­ 2/day with a mean value
2
to flow from this area towards the western and the soutern of 6.27 m­ /day. A transmissivity range of 0.36–13.47 m ­ 2/
parts of the study area. day and a mean value of 3.03 m ­ 2/day were reported in in
Though the specific capacity values obtained ranges an earlier work by Asante-Annor et al. (2018) in the area.
between 0.3 and 16.9 ­m3/day/m with a mean of 4.98 ­m3/ It is observed that most of the areas with high borehole
day/m, spatial consideration indicates almost the entire yields are also zoned within the areas delineated as the
area of having specific capacity ranging from 1 to 11 m ­ 3/ relatively high transmissive areas. Based on the Kransy’s
day/m (Fig. 7a). A similar range of specific capacity val- classification scheme, 33.26 ­km2 (1.4%) of the area has
ues of 1.1–18.32 m ­ 3/day/m was recorded by Asante-Annor very low transmissivity (0.1–1 ­m2/day) and, hence, can
(2018). Again, it is observed that the area with relatively only sustain withdrawals for local water supply with lim-
high specific capacity coincides with the high yield zone in ited consumption, 1502.01 ­km2 (62.1%) of the area has
Fig. 7b. Borehole yields are in the range of 8.5–427 L/min low transmissivity (1–10 ­m2/day) and can sustain smaller
within the study area and is averagely 92.05 L/min. Asante- withdrawals for local water supply (private consumption),
Annor et al. (2018) reported yields of 60–400 L/min with an whereas the remaining 881.73 k­ m2 (36.5%) has intermedi-
average of 134.60 L/min. Nontheless, it is worth knowing ate transmissivity (10–100 m ­ 2/day) and therefore suitable

13
Sustainable Water Resources Management (2022) 8:55 Page 11 of 15 55

Fig. 7  Overlay of spatial distribution of a specific capacity, b borehole yield and c transmissivity on geological formations in the study area

for withdrawals for local water supply (small communities of cokriging yielded equal accuracy as the weighted over-
and plants). lay, cokriging relatively outperformed weighted overlay in
each corresponding parametric combination. Also, it was
Groundwater potential of the area observed (Table 4) that increasing the number of hydrogeo-
logical parameters in the weighted overlay approach had no
Table 4 shows the accuracy obtained for each paramet- improvement on the prediction accuracy of the potential
ric combination in cokriging and the weighted overlay maps, as in the case of Maps 4 and 5.
approaches whilst Fig. 8a, b shows the potential maps gen- According to the CWSA borehole design guidelines for
erated. In all, the cokriging technique with the parametric small communities (CWSA 2010), the minimum yield of a
combination of yield, specific capacity, transmissivity and borehole to be considered for hand pump installation is 13.5
static water level produced the best accuracy of 66.7% whilst L/min while the minimum yield of a borehole to be consid-
its corresponding weighted overlay method had an accuracy ered for mechanisation is 85 L/min. However, depending
of 44%. Generally, with the exception of the parametric on a comprehensive assessment of existing hydrogeologi-
combination of yield, specific capacity, water strike and cal conditions, and an adequate technical evaluation of the
aquifer zone thickness (Map 2), in which the performance yield of available boreholes, lower yields of up to 10 L/min
and yields lower than 80 L/min may be considered for hand

13
55 Page 12 of 15 Sustainable Water Resources Management (2022) 8:55

pump installation and mechanization, respectively. Hence, cokriging and weighted overlay approaches, with prediction
in the creation and validation of the groundwater potential accuracies of 66.7% and 44%, respectively. It is observed
maps, areas with boreholes yields lower than 13.5 L/min that aquifers in the area are mainly semi-confined to con-
were generally considered to have very low potential, areas fined, which confirms earlier studies in the area by Buckley
with yields between 13.5 and 20 L/min were considered to (1986) and Gibrilla et al. (2010). Exploratory spatial sta-
have low potential, areas with yields between 20 and 50 L/ tistical analyses of the various hydrogeological parameters
min were considered to have moderate potential, those with revealed that availability of groundwater (borehole yield)
yields between 50 and 100 L/min were deemed to have high in the area highly correlates with specific capacity, but has
potential, and areas with yields > 100 L/min considered to moderate correlation with transmissivity as well as SWL.
be of very high potential. Other hydrogeological parameters such as overburden thick-
Thus, the groundwater potential of the area was gener- ness, fractured zone thickness, etc., however, showed weak
ally categorised based on the cokrigging output as ranging to no correlation with the borehole yield. This is consistent
from very low, through moderate-to-high potential (Fig. 8a). with results reported by Kanagaraj et al. (2019) in another
Spatial evaluation of this cokrigging output indicates that area underlained by crystalline rocks with a combination
2.78 ­km2 representing 0.1% of the area has very low ground- of fractured and weather aquifer systems like the current
water potential, 850.45 ­km2 representing 35.2% has low study area.
potential, 1054.62 k­ m2 making up 43.6% of the area has On the other hand, a similar groundwater potential deline-
moderate potential, and the remaining 21.1% with an area ation study by Nsiah et al. (2018) in the sedimentary ter-
coverage of 509.15 ­km2 has high potential. Similar to the rain of Nabogo basin showed borehole yields to be better
cokriging output groundwater potential map, the catego- correlated with regolith (overburden) thickness than SWL.
ries of groundwater potential of the area range from very This change may be attributed to the difference in geology
and nature of aquifers in the two areas. Also, the spatial
integration of hydrogeological parameters (i.e., borehole
Table 4  Accuracy obtained for generated maps
yield, regolith thickness, SWL and transmissivity) by Nsiah
Map number Validation accuracy (%) et al. (2018) using the weighted overlay approach produced a
Cokriging Weighted groundwater potential map of much higher prediction accu-
overlay racy than in this study, which may also be to the difference
in geology and aquifer nature in the areas.
Map 1 66.7 44
Aside the use of hydrogeological parameters, Gumma and
Map 2 33.3 33.3
Pavelic (2013) integrated surficial parameters (i.e., geomor-
Map 3 56 33.3
phology, geology, slope, drainage density, annual rainfall,
Map 4 – 33.3
land use/land cover, and soil type) using the weighted over-
Map 5 – 33.3
lay approach to generate the groundwater potential map of
Ghana, which had a good correlation with borehole yields.
low through moderate to high for the weighted overlay out- Though the study was on a country scale, the study catego-
put potential map (Fig. 8b). Thus, a total area of 9.01 k­ m2 rized the area into three zones of poor, moderate, and good
(0.4%) has very low potential, 907.73 ­km2 (37.6%) has low groundwater potential zones, which is consistent with the
potential, 1148.53 k­ m2 (47.5%) has moderate potential and findings of the current study in a more localized scale. Simi-
the remaining 351.73 k­ m2 (14.6%) has high potential per larly, the weighted overlay technique was a success in delin-
the overlay output. Geological evaluation of the generated eating groundwater potential zones in an area in India utiliz-
potential maps shows that the undifferentiated granite zones ing the surficial parameters (Mukherjee et al. 2012). Again,
have the highest groundwater potential ranging from low to the weighted overlay technique has been used to integrate a
high, whereas the biotite gneiss and leucogranites have low- combination of hydrogeological and surficial parameters for
to-moderate potential with the sericite schist and the Banket groundwater delineations in other studies (Kanagaraj et al.
formation having low potential. 2019; Al-Abadi et al. 2021, etc.), and it produced groundwa-
ter potential maps with prediction accuracy like this study.
Discussion Thus, it is observed that the parameters and the weighted
overlay method of integration used in this study have been
The study area, predominantly underlained by crystalline employed to successfully delineate groundwater poten-
intrusive granitic rocks of the Birimian system, is catego- tial zones in Ghana and other parts of the world, all of
rized into very low, low, moderate, and high groundwater which yielded similar results. Though the cokriging over-
potential zones by the integration of borehole yield, spe- lay approach produced maps of better prediction accuracy
cific capacity, water strike and aquifer zone thickness using than the weighted overlay approach for the same parametric

13
Sustainable Water Resources Management (2022) 8:55 Page 13 of 15 55

Fig. 8  Groundwater potential maps for the a cokriging and b weighted overlay approaches

combination, not much has been done on the method with has moderate potential and 351.73 k­ m2 (14.6%) has high
respect to groundwater potential delineation. potential.
Geologically, the groundwater potential within the undif-
ferentiated granitic terrains in the area range from low to
Conclusion high, the biotite gneiss and the leucogranites have low-to-
moderate potential whereas the sericite schist and the Banket
This study has comprehensively assessed the ground- formation in the area have low groundwater potential. The
water potential of the Assin municipalities by generating study, therefore, offers very practical information on delinea-
groundwater potential maps of the area using cokriging and tion of groundwater potential zones and is expected to guide
weighted overlay approaches in ArcGIS. It was observed that in effective development of groundwater resources in the
the cokriging technique produced maps with better predic- study area. It should, however, be noted that this study was
tion accuracy than the weighted overlay integration. Also, conducted in an area underlained by crystalline basement
increasing the number of hydrogeological parameters in the rocks; hence, there may be variations in the outputs when the
weighted overlay approach had no improvement on the pre- method is applied in areas underlained by different geology.
diction accuracy of the potential maps.
Generally, borehole depth in the area ranges from 15 to Acknowledgements Authors wish to thank Tullow Ghana for sup-
porting this research through a graduate student scholarship. Again,
80 m with an average of 52 m, overburden thickness ranges the authors acknowledge Community Water and Sanitation Agency,
from 9 to 18 m, the mean aquifer thickness is 19.96 m and Central Region, Ghana for making data available for the research.
the water strike from the ground level ranges from 12 to
62 m. Also, the yield of boreholes in the area ranges from Funding No funding was received for conducting this study.
9 to 427 L/min with a mean of 92 L/min and their recov-
ery rates after an hour are generally greater than 80%. Availability of data and materials The datasets generated during and/or
analysed during the current study are available from the corresponding
Additionally, the specific capacity values in the area range author on reasonable request.
from 0.3 to 16.9 m­ 3/day/m with a mean of 4.98 m3/day/m
whereas the transmissivity ranges from 0.1 to 28.2 m ­ 2/day Declarations
2
with a mean value of 6.27 ­m /day. Spatially, the ground-
water potential of the study area varies from low to high Conflict of interest The authors have no conflicts of interest to declare
with 850.45 k­ m2 (35.2%) of the area having low potential, that are relevant to the content of this article.
1054.62 ­km2 (43.6%) possessing moderate potential, and
509.15 ­km2 (21.1%) of the area having high potential based
on the cokriging potential map. The weighted overlay poten-
tial map, on the other hand, indicates that a total area of
907.73 ­km2 (37.6%) has low potential, 1148.53 k­ m2 (47.5%)

13
55 Page 14 of 15 Sustainable Water Resources Management (2022) 8:55

References Ganyaglo SY, Banoeng-Yakubo B, Osae S, Dampare SB, Fianko JR,

Bhuiyan MA (2010) Hydrochemical and isotopic characterisa-
tion of groundwater in Eastern Region of Ghana. Water Resour
Ahmadi SH, Sedghamiz A (2008) Application and evaluation of Krig-
Protect 2:199–208
ing and Cokriging methods on groundwater depth mapping. Envi-
Ganyaglo SY, Osae S, Dampare SB, Fianko JR, Bhuiyan MA, Gibrilla
ron Monit Assess 138:357–368
A, Osei J (2012) Preliminary groundwater quality assessment in
Akinlalu AA, Adegbuyiro A, Adiat KAN, Akeredolu BE, Lateef WY
the central region of Ghana. Environ Earth Sci 66(2):573–587
(2017) Application of multi-criteria decision analysis in prediction
Gibrilla A, Akiti TT, Osae S, Adomako D, Ganyaglo SY, Bam EPK,
of groundwater resources potential: a case of Oke-Ana, Ilesa Area
Hadisu A (2010) Origin of dissolve ions in groundwaters in the
Southwestern, Nigeria. NRIAG J Astron Geophys 6(1):184–200
Northern Densu River Basin of Ghana Using stable isotopes of
Al-Abadi AM, Fryar AE, Rasheed AA, Pradhan B (2021) Assessment
18O and 2H. J Water Resour Prot 2(12):1010–1019
of groundwater potential in terms of the availability and qual-
Gibrilla A, Bam EKP, Adomako D, Ganyaglo S, Osae S, Akiti TT,
ity of the resource: a case study from Iraq. Environ Earth Sci
Kebede S, Achoribo E, Ahialey E, Ayanu G, Agyeman EK (2011)
80(12):1–22
Application of Water Quality Index (WQI) and multivariate analy-
Andualem TG, Demeke GG (2019) Groundwater potential assess-
sis for groundwater quality assessment of the Birimian and Cape
ment using GIS and remote sensing: a case study of Guna Tana
Coast Granitoid Complex: Densu River Basin of Ghana. Water
Landscape, Upper Blue Nile Basin Ethiopia. J Hydrol Reg Stud
Qual Expo Health 3(2):63–78
24:100610
Giraldo R, Herrera L, Leiva V (2020) Cokriging prediction using as
Anteneh Z, Alemu MM, Bawoke GK 1994 etnet, Kehali A, Fenta M,
secondary variable a functional random field with application in
Desta MT (2021) Appraising groundwater potential zones using
environmental pollution. Mathematics 8(8):1305
geospatial and multi-criteria decision analysis (MCDA) tech-
Gogu R, Carabin G, Hallet V, Peters V, Dassargues A (2001) GIS-
niques in Andasa-Tul Watershed, Upper Blue Nile basin, Ethiopia.
Based hydrogeological databases and groundwater modelling.
https://​doi.​org/​10.​21203/​rs.3.​rs-​309494/​v1
Hydrogeol J 9(6):555–569
Armah TK (2000) Groundwater salinization in parts of Southern
Goodchild MF (1993) The state of GIS for environmental problem-
Ghana. In: Oliver S (ed) Groundwater: past achievements and
solving. In: Goodchild MF, Parks BO, Steyaert LT (eds) Environ-
future challenges, (Proceedings of IAH Congress on Groundwater,
mental modeling with GIS. Oxford University Press, New York,
26 November 2000). A.A. Balkema Publication, Rotterdam. Pp
pp 8–15
445–449
Goovaerts P (1997) Geostatistics for natural resources evaluation.
Arulbalaji P, Padmalal D, Sreelash K (2019) GIS and AHP techniques-
Oxford University Press on Demand
based delineation of groundwater potential zones: a case study
GSA (Geological Survey Authority) (2009) Geological Map of Ghana
from Southern Western Ghats India. Sci Rep 9(1):1–17
1:1000000, GSD, Accra, Ghana
Asante-Annor A, Acquah J, Ansah E (2018) Hydrogeological and
GSD (Geological Survey Department) (2009) Geological Map of
hydrochemical assessment of basin granitoids in Assin and Bre-
Ghana 1:1000000, GSD, Accra, Ghana
man Districts of Ghana. J Geosci Environ Protect 6(9):31–57
GSS (Ghana Statistical Service) (2012) Population and Housing Cen-
Bashe BB (2017) Groundwater potential mapping using remote sensing
sus. Ghana Statistical Service, Accra, Ghana
and gis in Rift Valley Lakes Basin, Weito Sub Basin Ethiopia. Int
Gumma MK, Pavelic P (2013) Mapping of groundwater potential
J Sci Eng Res 8(2):43–50
zones across Ghana using remote sensing, geographic infor-
Buckley DK (1986) Report on advisory visit to water aid projects in
mation systems, and spatial modeling. Environ Monit Assess
Ghana. Unpublished Report, British Geological Survey, Hydro-
185(4):3561–3579
geology Research Group, Wallingford
Hoeksema RJ, Clapp RB, Thomas AL, Hunley AE, Farrow ND,
Chowdhury A, Jha MK, Chowdary VM, Mal BC (2009) Integrated
Dearstone KC (1989) Cokriging model for estimation of water
remote sensing and GIS-based approach for assessing groundwa-
table elevation. Water Resour Res 25:429–438
ter potential in West Medinipur District, West Bengal India. Int J
Hooshmand A, Delgh M, Izadi A, Amadali KA (2011) Application
Remote Sens 30(1):231–250
of Kriging and Cokriging in spatial estimation of groundwater
CWSA (Community Water and Sanitation Agency) (2010) Small Com-
quality parameters. Afr J Agric Res 6(14):3402–3408
munity Sector Guidelines. CWSA Publication, Accra
Hu K, Huang Y, Li H, Li B, Chen D, White RE (2005) Spatial vari-
Deutsch CV, Journel AG (1992) Geostatistical Software Library and
ability of shallow groundwater level, electrical conductivity and
User’s Guide. New York, 119(147)
nitrate concentration, and risk assessment of nitrate contamina-
Dickson KB, Benneh G, Essah RR (1988) A new geography of Ghana.
tion in North China Plain. Environ Int 31(6):896–903
Longman Group Ltd., London, p 34
Hussein AA, Govindu V, Nigusse AGM (2017) Evaluation of
Edet AE, Okereke CS (1997) Assessment of hydrogeological con-
groundwater potential using geospatial techniques. Appl Water
ditions in basement aquifers of the Precambrian Oban Massif,
Sci 7(5):2447–2461
Southeastern Nigeria. J Appl Geophys 36(4):195–204
Israil M, Al-Hadithi M, Singhal DC (2006) Application of a resistiv-
Eldeiry AA, Garcia LA (2012) Evaluating the performance of
ity survey and geographical information system (GIS) analysis
ordinary kriging in mapping soil salinity. J Irrig Drain Eng
for hydrogeological zoning of a piedmont area, Himalayan Foot-
138(12):1046–1059
hill Region India. Hydrogeol J 14(5):753–759
ESRI (Environmental Systems Research Institute) (2016). ArcMap.
Jaiswal RK, Mukherjee S, Krishnamurthy J, Saxena R (2003) Role
How Kriging works. Retrieved from https://​www.​deskt​op.​arc-
of remote sensing and GIS techniques for generation of ground-
gis.​com/​en/​arcmap/​1.0.​3/​tools/​3d-​analy​st-​toolb​ox/​how-​krigi​ng-​
water prospect zones towards rural development—an approach.
works.​htm#. Accessed on 09 Jul 2021
Int J Remote Sens 24(5):993–1008
Ewusi A, Kuma JS (2011) Calibration of shallow borehole drilling sites
Jha MK, Peiffer S (2006) Applications of remote sensing and GIS
using the electrical resistivity imaging technique in the Granitoids
technologies in groundwater hydrology: past, present and future.
of Central Region Ghana. Nat Resour Res 20(1):57–63
BayCEER, Bayreuth
FAO (Food and Agriculture Organization of the United Nations). 2002.
Jhariya DC, Kumar T, Gobinath M, Diwan P, Kishore N (2016)
Project Intégré Keita. Rapport Terminal du Project. Project GCP/
Assessment of groundwater potential zone using remote
NER/032/ITA. Rome, FAO

13
Sustainable Water Resources Management (2022) 8:55 Page 15 of 15 55

sensing, GIS and multi criteria decision analysis techniques. J Shahid S, Nath S, Roy J (2000) Groundwater potential modelling in a
Geol Soc India 88(4):481–492 soft rock area using a GIS. Int J Remote Sens 21(9):1919–1924
Kanagaraj G, Suganthi S, Elango L, Magesh NS (2019) Assessment Sikdar PK, Chakraborty S, Adhya E, Paul PK (2004) Land use/land
of groundwater potential Zones in Vellore district, Tamil Nadu, cover changes and groundwater potential zoning in and around
India using geospatial techniques. Earth Sci Inf 12(2):211–223 Raniganj Coal Mining Area, Bardhaman District, West Bengal—a
Kortatsi BK (1994) Groundwater utilization in Ghana. IAHS Publi- GIS and remote sensing approach. J Spat Hydrol 4(2):1–24
cations-Series of Proceedings and Reports-Intern Assoc Hydro- Sreedevi PD, Subrahmanyam K, Ahmed S (2005) Integrated approach
logical Sciences, 222, 149–156 for delineating potential zones to explore for groundwater in the
Krásný J (1993) Classification of transmissivity magnitude and vari- Pageru River Basin, Cuddapah District, Andhra Pradesh India.
ation. Groundwater 31(2):230–236 Hydrogeol J 13(3):534–543
Krishnamurthy J, Mani A, Jayaraman V, Manivel M (2000) Ground- Srinivasa Rao Y, Jugran DK (2003) Delineation of groundwater
water resources development in hard rock terrane-an approach potential zones and zones of groundwater quality suitable for
using remote sensing and GIS techniques. Int J Appl Earth domestic purposes using remote sensing and GIS. Hydrol Sci J
Observ Geo-Inform 2(3–4):204–215 48(5):821–833
Kumar S, Machiwal D, Parmar BS (2019) A parsimonious approach Srivastava PK, Bhattacharya AK (2006) Groundwater assessment
to delineating groundwater potential zones using geospatial through an integrated approach using remote sensing, GIS and
modeling and multicriteria decision analysis techniques under resistivity techniques: a case study from a hard rock terrane. Int J
limited data availability condition. Eng Rep 1(5):2073 Remote Sens 27(20):4599–4620
Magesh NS, Chandrasekar N, Soundranayagam JP (2012) Deline- Stafford DB (1991) Civil engineering applications of remote sensing
ation of groundwater potential Zones in Theni District, Tamil and geographic information systems. In: A conference, (Proceed-
Nadu, using remote sensing GIS and MIF Techniques. Geosci ings of the Second National Specialty Conference, 14–16 May
Front 3(2):189–196 1991). American Society of Civil Engineers
Mallick J, Khan RA, Ahmed M, Alqadhi SD, Alsubih M, Falqi I, Theis CV (1935) The relation between the lowering of the piezo-
Hasan MA (2019) Modeling groundwater potential zone in a metric surface and the rate and duration of discharge of a well
Semi-Arid Region of Aseer using fuzzy-AHP and geo-informa- using ground-water storage. EOS Trans Am Geophys Union
tion techniques. Water 11(12):2656 16(2):519–524
Mohammed-Aslam MA, Kondoh A, Rafeekh PM, Manoharan Todd DK (1980) Groundwater hydrology. Wiley, New York
AN (2010) Evaluating groundwater potential of a hard-rock UNESCO (United Nations Educational, Scientific and Cultural Organi-
aquifer using remote sensing and geophysics. J Spat Hydrol zation) (2012) World’s Groundwater Resources are Suffering from
10(1):76–88 Poor Governance. UNESCO Natural Sciences Sector News. Paris,
Mukherjee P, Singh CK, Mukherjee S (2012) Delineation of ground- UNESCO
water potential zones in Arid Region of India—a remote sensing van Tonder G, Kunstmann H, Xu Y, Fourie F (2000) Estimation of
and GIS approach. Water Resour Manag 26(9):2643–2672 the sustainable yield of a borehole including boundary informa-
Murthy KSR (2000) Groundwater potential in a semi-arid region of tion, drawdown derivatives and uncertainty propagation. In: Cali-
Andhra Pradesh—a geographical information system approach. bration and Reliability in Groundwater Modelling: Coping with
Int J Remote Sens 21(9):1867–1884 Uncertainty, (Proceedings of the ModelCare'99 Conference Held
NRC (National Research Council) (1996) Rock fractures and fluid in Zurich, Switzerland, 20–23 September 1999) IAHS Publica-
flow: contemporary understanding and applications. National tions 265: 367
Academies Press, Washington Waikar ML, Nilawar AP (2014) Identification of groundwater potential
Nsiah E, Appiah-Adjei EK, Adjei KA (2018) Hydrogeological deline- zone using remote sensing and GIS technique. Int J Innov Res Sci
ation of groundwater potential zones in the Nabogo Basin, Ghana. Eng Technol 3(5):12163–12174
J Afr Earth Sc 143:1–9 WRI (World Resources Institute) (1992) World Resources, 1992–1993.
Oh HJ, Kim YS, Choi JK, Park E, Lee S (2011) GIS mapping of Oxford University Press, New York
regional probabilistic groundwater potential in the area of Pohang Xu Y, Usher B (2006) Issues of groundwater pollution in Africa. In:
City Korea. J Hydrol 399(3–4):158–172 By Y (ed) Groundwater pollution in Africa. Taylor & Francis,
Pinto D, Shrestha S, Babel MS, Ninsawat S (2017) Delineation of London, pp 3–9
groundwater potential zones in the Comoro Watershed, Timor Yalcïn E (2005) Cokriging and its effect on the estimation precision. J
Leste using GIS, remote sensing and analytic hierarchy process South Afr Inst Min Metall 105(4):223–228
(AHP) technique. Appl Water Sci 7(1):503–519 Yeh HF, Lee CH, Hsu KC, Chang PH (2009) GIS for the assess-
Prackley S (1984) The 30-well project. Internal Report, Catholic Dio- ment of the groundwater recharge potential zone. Environ Geol
cese of Accra, Accra 58(1):185–195
Preeja KR, Joseph S, Thomas J, Vijith H (2011) Identification of Yifru BA, Mitiku DB, Tolera MB, Chang SW, Chung IM (2020)
groundwater potential zones of a Tropical River Basin (Kerala, Groundwater potential mapping using SWAT and GIS-based
India) using remote sensing and GIS techniques. J Indian Soc multi-criteria decision analysis. KSCE J Civ Eng 24(8):2546–2559
Remote Sens 39(1):83–94 Zare CM, Zare CA, Malekian A, Bagheri R, Vesali SA (2014) Evalu-
Rao NS (2003) Groundwater prospecting and management in an agro- ation of different cokriging methods for rainfall estimation in
based rural environment of crystalline terrane of India. Environ Arid Regions (Central Kavir Basin in Iran). Desert (BIABAN)
Geol 43(4):419–431 19(1):1–9
Sander P, Chesley MM, Minor TB (1996) Groundwater assessment
using remote sensing and GIS in a rural groundwater project in Publisher's Note Springer Nature remains neutral with regard to
Ghana: lessons learned. Hydrogeol J 4(3):40–49 jurisdictional claims in published maps and institutional affiliations.
Sener E, Davraz A, Ozcelik M (2005) An integration of GIS and
remote sensing in groundwater investigations: a case study in
Burdur Turkey. Hydrogeol J 13(5–6):826–834
Shahid S, Nath SK (2002) GIS integration of remote sensing and
electrical sounding data for hydrogeological exploration. J Spat
Hydrol 2(1)

13