The study area form part of two major litho-tectonic settings and have experienced several tectonic activities. It is situated on the southern coast of Ghana and forms part of the Kibi-Winneba volcanic belt, belonging to the Paleoproterozoic Birimian Supergroup of the West African Craton (WAC) and the Dahomeyan basement rocks of the Dahomeyide belt, Southeastern Ghana, belonging to the Neoproterozoic Trans-Saharan Shear Zone, West Africa. The Birimian terrane is a younger crust formed approximately 2.2 - 2.1 Ga during the Paleoproterozoic Era ([11]). The Birimian rocks were changed as a result of the Eburnean orogenic event, which is the main tectono-magmatic orogenic event responsible for the re-structuring of the Birimian area. The Birimian Supergroup is widely emplaced across Ghana and the entire West African subregion. It is made up of metamorphosed volcanic and sedimentary units intruded by granitic rocks. The Birimian terrane is part of the eastern segment of the West African Craton known as the Leo-Man Shield ([21]). The rocks are configured in a belt-basin geometry, thus, metavolcanic belts known as the greenstones are separated by metasedimentary basins. The Kibi-Winneba belt is a southeastern Birimian belts in Ghana covering parts of the Winneba area. The belt is oriented in the northeast-southwest direction, which matches the regional structural trend of Birimian terranes in West-Africa ([7]). The belt covers parts of the Eastern and Central Regions of Ghana, with the Kibi municipality to the north and the coastal town of Winneba to the south. Lithologically, the Kibi-Winneba belt contains metavolcanic rocks, metasedimentary units and intrusive granitoids. The metavolcanic rocks are predominantly mafic (basalts) to intermediate (andesites) in composition, while the metasedimentary units consist mainly of greywacke, shale and volcaniclastics ([19]). Granitic intrusives are common all over the Birimian terrane. The granitic rocks in the Kibi-Winneba belt show volcanic-arc geochemical signatures. They evolved from crustal melting as a result intensive thermal activity associated with the Eburnean orogenic event ([3]). They are composed of tonalites and granodiorite intruding the metamorphosed volcanic and sedimentary units. Structurally, the rocks of the Kibi-Winneba belt are characterized by faults, folds and shear zones formed during Eburnean deformational event. The geological structural configuration is a factor that controls mineralization and fluid accumulation in the belt. [6] stated that the main areas of mineralization in the belt are tectonized shear contacts and fracture networks. The belt hosts quartz veins and sulphide gold mineralization similar to other Birimian belts. Mesothermal gold deposits occur with arsenopyrite and pyrite as pathfinders formed from hydrothermal fluids accumulated in shear contacts ([9]).

The Trans-Saharan Shear Zone is an important Pan-African tectonic belts in West Africa. It runs in a broadly north-northeast to south-southwest direction along the eastern margins of Ghana, Togo and Benin through to southwestern Nigeria. This belt marks the collisional zone formed during the Pan-African orogeny, a period of continental convergence that occurred around 600-550Ma during the late Neoproterozoic Era ([1]; [5]). The southern part of the Pan-African Trans-Saharan Shear Zone which occur in Ghana is known as the Dahomeyide belt. The Dahomeyide belt resulted from the collision of existing rocks of the West African Craton (WAC) in the west, with the rocks of the Nigeria-Benin Shield to the east. This resulted to a westward subduction of oceanic crust and the emplacement of high-grade metamorphic and granitic rocks along the collisional margin ([4]). The belt is structurally divided into three major litho-tectonic zones that are characterized by deformation and metamorphic conditions from west to east. The western zone, made up of low-grade deformed marginal sediments of the Voltaian Basin consisting of metasedimentary rocks which are folded and faulted. The central suture zone, hosts high and ultra-high-pressure rocks of mantle origin including serpentinites, peridotites, amphibolies and eclogites. Also present are highly metamorphosed rocks of granitic origin made up of nepheline syenite gneisses, garnet gneisses and sericite gneisses ([4]; [8]). The eastern zone is covered by high-grade gneisses, migmatites and granulites of the rejuvenated passive margin rocks of the Nigeria-Benin Shield, that experienced deformation and high-grade metamorphism during the collision.

The study area (Figure 1) is Winneba and its environs in the Effutu Municipality, Central Region of Ghana. The town Winneba lies approximately 65 km southwest of Accra, the national capital and 50 km east of Cape Coast, the Regional Capital. Geographically, the study area is bounded by latitudes 5˚18'N to 5˚28'N and longitudes 0˚32'W to 0˚42'W covering an area of roughly 200 km² that encompasses some coastal towns with immediate inland communities. Geologically, the area falls within the transition between the Kibi-Winneba belt of the Birimian Supergroup and the Precambrian basement rocks of the Dahomeyide belt with a Quaternary coastal sedimentary cover. Access to the area is primarily via the “N1” coastal highway connecting the national capital, Accra to Cape Coast, the regional capital with secondary roads extending inland toward the surrounding communities. The Winneba area is generally low-lying, with elevations ranging from sea level to approximately 80 m above mean sea level. The terrain is predominantly flat to gently undulating, becoming more dissected northward where resistant metamorphic and granitoid lithologies exert greater topographic control. The coastline to the south of the study area is irregular and characterized by sandy beaches. Extensive areas of coastal scrub, thicker vegetation and secondary bushes cover much of the study area, significantly limiting geological exposures and accessibility to the bedrock. This dense vegetation cover, combined with thin lateritic soils and a coastal sediment blanket, substantially restricts the effectiveness of conventional surface geological mapping and provides a primary justification for the application of aeromagnetic methods in this study. Settlement patterns in the study area reflect a coastal-inland dichotomy, with dense nucleated settlements along the coastline and smaller dispersed farming communities distributed across the inland portions of the study area.

Figure 1.The geological map of the study area.

Field surveyed aero-magnetics data was used for this research. The data used for this research was obtain from the Geological Survey Authority of Ghana. It was acquired in the year 2000 by High Sense Geophysics Incorporation, Canada. The initial field survey (Figure 2) was designed on equal survey line intervals to form a linearly spaced grid of 400 m at an angle of 315˚ with the supporting tie lines spaced at 5000 m intervals at 045˚ orientation. The data was acquired along the planned survey lines across the strike of the various lithologies using a proton precession magnetometer onboard an airplane flying at a nominal terrane clearance of 80 m. A total area of 200 km² was covered by the survey flight at a Nominal Airspeed 100 km/hr. Flight drift and tilt which are data errors associated with the flight path were corrected. These flight discrepancies are as a result of cross wind and wrong flight maneuvering. The final Magnetic Field Intensity (TMFI) comprising of the regional and residual fields were measured for further processing.

Figure 2.Aeromagnetic field survey grid lines.

Magnetic data preprocessing which include a number of data reduction procedures were applied to the acquired data using the Geosoft Oasis Montaj software. The data was edited to remove data spike values which are inconsistent with the data distribution. Magnetic field noise originating from cultural feature were also corrected. The resulting data contained only relevant magnetic field information from the rock’s distribution in the study area. Further processing steps applied are data checking, diurnal changes removal, tie line leveling, geomagnetic reference field correction and micro leveling. The data was gridded using the minimum curvature estimation grid method with a grid cell size 100 m considering a one-quarter value from a field survey interval of 400 m. The result was a Total Field Intensity grid map showing the variation of magnetic field across the study area. Using an inclination and a declination values of −13˚ and −6˚ respectively at the time the field survey was conducted, the total field intensity was reduced to the equator to produce the true anomalous distribution in the study area. To enhance various geological features, interpretative grids such as Analytical Signal grid, Horizontal derivative, Vertical derivative, tilt derivative and their corresponding shaded intensity grid maps were produced. The robust CET Grid analyses and the Advanced Source Parameter Imaging technique were applied to analyze for lineaments and calculate relative depths to magnetic sources. The CET Grid analyses combine the edge detection filters to develop an integrated lineament system devoid of human influence.

The Advanced CET Grid Analysis was used to map linear features using the edges detection derivatives of the total magnetic intensity grid as input. The analyses are done automatically by delineating all possible lineament in all directions across the basement. The processing steps include Lineation detection, Amplitude thresholding and Lineament Vectorisation. Lineation detection delineate geological edges in field data irrespective of the field intensity orientation. The Amplitude thresholding procedure involve finding and mapping the linear thresholds which are the highest amplitudes in the grid. Mapped directional lines are then placed on the delineated linear thresholds through the process of Lineament Vectorisation. This automated lineament mapping tool is very robust such that all minor lineaments are approximated to aid in interpretation. Geological edge locations are indicative of magnetic intensities associated with magmatic intrusions which normally contain higher magnetic minerals. The CET Grid Analysis method maps geological edges on local magnetic intensity maximums and connects the output to symmetry locations within its immediate area. Lower magnetic field intensity measurements are suppressed while higher magnetic field intensity measurements are enhanced using the amplitude thresholding filter. The amplitude thresholding filter apply a maximal enhancement filter (NMF) to evaluate the data to determine the highest regions. The final delineated edges are emplaced using a skeletonization filter. This filter repeatedly joins boundary cells by connecting grid cell with mapped lineaments. The traditional edges detection methods which are horizontal, vertical and tilt derivative grids were used as an input data for the CET Grid Analysis. The delineated lineaments were further interpreted as faults, dykes and lithologic contacts. Magnetic intensity maximums of horizontal derivative grids occur above geological litho-contacts while that of vertical derivative grids occur at fractured contacts such as faults. These same edges are mapped by marking the contacts of magnetic intensity highs and lows by the tilt derivative filter. The horizontal derivative of a magnetic field intensity is calculated from the equation:

where “HD” is the horizontal derivative, ( d M d X ) is the “x” derivative of the measured field “M” and ( d M d Y ) is the “y” derivative of the measured field “M”.

The above equation gives the highest location above magnetic anomalies where the regional magnetic field is vertical, the magnetizations are vertical, the contacts are vertical, the contacts are isolated and the sources are thicker ([18]). The challenge with vertical derivative filter is magnetic intensity peak shifts and the creation of additional peaks which has the capacity to mask true contacts. The vertical derivative is calculated by Equations (2):

where d M / d Z is the derivative of the magnetic field “M” in the Z-direction. Miller and Singh (1994) also showed that, the tilt derivative crosses through zero contours at or near the edge of a vertical-sided source and is negative outside the source region. The tilt derivative is therefore very effective in allowing anomalies to be traced out along strike. According to [24], the generalized tilt derivative equation is given as:

where VD is the first vertical derivative of the measured field (M) and HD is the horizontal derivative. Thus, the tilt derivative is the ratio of the vertical derivative to the absolute amplitude of the total horizontal derivative of a magnetic field. Substituting Equations (1) and (2) in Equation (3) gives Equation (4);

Thurston[23] theorized the mathematical basis for a spatial depth estimation method known as the Source Parameter Imaging (SPI). It is an automatic estimation procedure which require minimal human interface. The method uses the local wavenumber of the analytic signal grid to determine peak for depth value estimation. The analytical signal grid is calculated from the total magnetic intensity field grid. The rate of change of the local wavenumber spatially, is associated with the spatial distribution of magnetic anomalies, anomaly geometry and their depth of occurrence. The procedure maps peak positions from the analytical signal grid using higher wavenumbers over magnetic sources. the calculated peak value is inversely proportional to the depth interval to the anomalous body ([23]). The depths measurements are then calculated from the local wavenumber at each grid vertices without prior information about the anomalies source locations, producing a spatially isolated relative depth grid map. The mathematical foundation of the ASPI is based on the relationship between the local wavenumber “k” and the sources depth “z”, given by;

where “n” is the structural index resulting from the shape of the anomalous source body ([20]). A Structural index values of 0.5, 1, and 2 correspond to geological contacts, dykes and pipe type geometric shapes bodies respectively. A structural index value 1 was used for the SPI analyses to model depth solutions along higher magnitude such as dykes. The ASPI technique is efficient for estimating depths to the magnetic source from spatial gridded data than distinct line field surveys ([10]). The ASPI Method is able to estimate precise depths without recourse to the direction of the inducing magnetic field which is advantageous over the traditional depth estimation methods. The purely induced magnetization concept is rooted in self-directional magnetization and the use of the analytic signal grid as the basal grid for depth estimation by the ASPI Method fulfill the fundamental basis for this theory ([16]). Comparing to the traditional derivative-based depth to magnetic sources estimation methods, the ASPI Method is not affected the inclination and declination of the total magnetic field making it a robust technique for mapping spatial depths.

Two-dimension anomaly models along the field survey lines were produced from the dataset (Figure 3). This was to assess the changes in magnetic intensities with associated anomalies in higher and lower intensities. An interactive bilinear method was used to model the field intensities along a y-axis with the field survey intervals on x-axis. The data was automatically inverted to produce a 2D-model for analyses. An Iterative inverse modelling approach was applied to model magnetic anomalies along survey lines using the Oasis Montaj software. This technique uses an automated routine to determine the adjustment to be made to the model parameters. The survey lines L4, L7, L9, and L14 were selected arbitrary to distribute the model profiles approximately equally across the study area. The Input parameters were the magnetic inclination, declination and the magnetic field background value at the time the data was taken. A magnetic field intensity data along a particular survey line is then chosen for modelling. Using the background field intensity value as an initial modelling parameter value, the inversion program conducts a number of iterations to determine the magnetic anomalies along the survey line. With a threshold anomalous values of zero (0) nanotesla, the modelled magnetic anomaly values range between +/− 250 nT. Figure 3(a) shows a 2D profile model along survey line “L4”. The line covers a survey interval of 7341 m. A high magnetic anomaly was modeled along the entire stretch of this survey line. The highest anomaly observed was 160 nT which occur at the survey location 3500 m. The modelled anomaly profile dips at locations 700 m, 2500 m, 3600 m with positive anomaly values of approximately 1 nT, 40 nT and 80 nT respectively. Figure 3(b) shows the modelled profile line “L7” along a 6510 m interval. The first 2000 m continuous interval and the last 3510 m survey length showed an intermediate high magnetic anomaly of approximately 100 nT. An extremely low anomaly of −200 nT occurring at location 2700 m is followed by a high anomaly of 230 nT at survey location 2900 m. There are established geological contact at the zero (0) anomaly boundary from higher anomaly to lower anomaly and vice versa. The modelled profile, Figure 3(c) for line “L9”, also demonstrated continuous intermediate high anomaly intensities along most of the survey length. The 7341 m line interval showed a middle lower anomaly, an intermediate and a high anomaly with measured field anomalies ranging from −150 nT to 150 nT. The highest anomaly value occurs at survey location 110 m with a measured positive anomaly value of 150 nT. The change in anomaly direction from high to low at location 3600 m boundary and low to high at location 3950 m boundary, shows a geological contact between two rock types in the study area. Figure 3(d) shows the changes in magnetic anomalies values along survey line “L14”. The survey line covers a 2652 m interval and shows a continuous high anomaly at both ends of the modeled profile with a lower magnetic anomaly in between them. At an interval range of 0 - 2250 m, a high positive anomaly value of 1 - 100 nT is recorded.

This is followed by a low anomaly of −100 nT at survey location 2300 m. A litho-contact occurs at the various transitions from high to low and low to high anomalies positions. A review of the modeled anomaly distribution along the profiles to the geology of the Winneba area, indicates the observed high magnetic anomaly areas correlates with the onshore solid basement rocks of the Dahomeyide belt which are mostly granitic gneisses with higher magnetic minerals. The measured lower anomalies corresponded to areas occupied by the overlying metasedimentary rocks of the Birimian Supergroup which are mostly meta-arenites and meta-argillites. The modeled intermediate anomalies lie between metavolcanics and metasedimentary rocks. Their magnetic anomaly signature shows a mixture of metavolcanics and meta-clastic rocks of the southeastern Dahomeyide and the Birimian rocks, hence, are interpreted as volcaniclastics. Theoretical knowledge of the study area shows geological contacts between the metavolcanics, volcaniclastics and the metasedimentary rocks.

Figure 3.2D profile models(a) Line “L4” (b) Line “L7” (c) Line “L9” (d) Line “L14”.

3.2.1. Total Magnetic Intensity Grid (TMI)

A total magnetic intensity grid (Figure 4) was produced from the dataset. The induced magnetization of rocks is characterized by the remnant magnetization and the field magnetization from the earth’s core. This include a residual and regional magnetic field intensity relating to the total magnetic field intensity. This shows the lateral distribution of the magnetic intensities in the study area. Higher intensities are indicated in red to magenta colours while lower magnetic intensities are shown in blue to green colours. The variations in the magnetic intensities is associated with the magnetization of the study area rocks. The position of Ghana is closer to the geographic equator and hence, the measured magnetic field intensities are overturned and shifted from their causative rocks, causing higher anomalies in blue colour to appear as lower anomalies and vice versa. As a result of this natural inversion of the magnetic field anomaly intensities, the crustal magnetization which are very strong occur to the southeast while weaker magnetizations occur to the northeast of the study area. The magnetic intensities measured ranges from 3 nT to approximately 125 nT. The observed magnetic intensities are from the differences in remnant magnetization of the overlying meta-sediments and the highly magnetic basement rocks.

Figure 4.Total magnetic intensity (TMI) grid map of the study area.

3.2.2. Analytical Signal Grid (AS)

Figure 5. Analytical signal grid maps of the study area.

The total magnetic field intensities grid was further processed using the analytical signal filter (Figure 5) to show the lithological variation in the study area. This was to position the observed field magnetic anomalies directly on their subsurface magnetic rocks. The analytical signal filter is well suited for eliminating the disparities in magnetic anomaly position within lower geographical latitudes. The location of Ghana makes it an ideal filter for identifying exact magnetic anomaly position for interpretation. The observed magnetic anomaly intensities range from −3 nT to 53 nT. Higher magnetic anomaly intensities occur as lineaments throughout the study area and are related to intrusive volcanic dykes partly associated with magmatic igneous rocks of the Birimian tectonic settings. These linearized bodies occurring as dykes labelled “D1” to “D5” are oriented in the northeast-southwest direction. Their trend is consistent with the general lineament trend of the Birimian tectonic settings. The magmatic igneous rocks are shown as patches labelled “V1” to “V5” of high magnetic anomaly intensities in magenta colours. To the south, toward the sea are sedimentary cover rocks indicated in lower magnetic anomaly intensities in blue colour. These are mostly particulate sand and silt size rocks along the shore line. Some intermediate signatures in green colour are scattered throughout the study area mostly around the higher magnetic anomaly intensity metavolcanic rocks. It is related to the metamorphosed volcaniclastics rocks accumulated from volcanic ash and the breakdown of the metavolcanics in the study area. The analytical signal anomaly intensity distribution was effective in distinguishing the rock types and also revealed deep seated magnetizations which could not be assessed on the surface.

3.2.3. CET Analyses

Figure 6.Mapped edge detection grid maps of the study area (a) Horizontal derivative (b) Vertical derivative (c) Tilt derivative (d) Composite lineaments map.

The study used the robust center for exploration targeting analyses to map out linearized structures in the research area. The method includes the Lineation detection, Amplitude thresholding and lineation vectorisation which assesses the survey area in smaller grid cells to locate magnetic intensities ridges to be mapped. These filters were applied to the edge detection method such as the horizontal, vertical and tilt derivatives (Figure 6). The edge detection methods initially derive the derivative of the absolute magnetic field anomalies to enhance lineaments for the CET analyses. The lineation detection filter place lines on the edges of the enhanced linear features automatically.

A threshold is set for the amplitude of the derivative derived field to limit the number and extent of the final lineaments to be considered from the individual edge detection methods. Through a process of vectorization, lines are placed on all possible linearized geological structures for further analyses. The CET analyses map out lineament at the maximum field location of the horizontal and vertical derivatives but at the maximum-minimum field boundary of the tilt derivative grid. The mapped individual lineation is combined by superposition to establish the true lineaments. The combined mapped lineaments from the CET Grid analyses were positioned closer together. The concept of [15] was used to establish the most reliable contact locations from the combined lineaments: 1) a true contact is established at locations where the three mapped lineaments is superimposed one another 2) possible contact locations were mapped horizontal derivative lineament is isolated from that of the vertical derivative and 3) at parallel contacts, the true contact position is the vertical derivative contact and not the tilt derivative contact ([15]). The first assumption provides the best contact positions which were achieved by tracing the final lineament as a layer with ArcMap of the ArcGIS software. The second condition satisfies the assumptions of the horizontal derivative theory ([18]).

3.2.4. Mapped Basement Geological Structure

The basement geological structural map of the study area is shown in Figure 7. The mapped geological structures include linearized and curvilinear structures. The linearized structures are probably faulting and dykes-like intrusion of volcanic sources while the curvilinear structure are geological boundaries which distinguishes the various rock types in the area. The area is covered by three rock types which include metasedimentary rocks to the west, the volcaniclastics which are centrally located and the metavolcanics which occupy the east of the study area. The two major litho-contacts are labelled “L1” and “L2”. The contact “L1” Separate the metasediments and volcaniclastics and the contact “T2” separate the volcaniclastics from the metavolcanics. The linearized structures are interpreted as dykes associated with various volcanic intrusions in the study area. They were grouped into major and minor dykes based on their extent, orientation and size of deposition. The major dyke labelled “D1” run through the entire study area in a northeast-southwest orientation. The minor dykes labelled “D2” to “D8” are associated with some of the major volcanic intrusions in the study area. The mapped lineaments are generally oriented in the northeast-southwest direction which is consistent with the structural trends in the Birimian Supergroup of Ghana. The underlying geological structure revealed a highly tectonized basement characterized by extensive shearing associated with basement dykes and volcanic intrusions.

Figure 7.Basement structural map of the study area showing mapped lineaments and litho-contacts.

The relative spatial depth solution values plot (Figure 8) shows a depth range of less than 150 m to >300 m. The study area showed a varied depth distribution. The Source Parameter Imaging (SPI) technique estimates depths to magnetic-source across an area. To estimate depths to the crystalline basement, the total magnetic field intensity (TMI) grid reduced to pole was separated into a residual and regional field intensities. The SPI depths were then estimated using the regional magnetic field intensity grid. The estimated depths to magnetic sources in this geological setting using the regional magnetic field intensities were interpreted as the depths to the crystalline basement or overburden thickness because they corresponded to previous overburden drilling depths within the study area. Depth to the basement crystalline rocks increases from the northwest to the southeast, as one approaches the shoreline. This is demonstrated by the continual variation in depth solution value colours from blue to red. Deeper depths (>300 m) are shown in red depth solutions while much shallower depth (less than 150 m) solutions are shown in blue solution. The intermediate depth (150 m - 300 m) solution occurs in both yellow and green colours. The cumulative average range depth measure to the underlying crystalline basement rocks across the study area is between 150 m to 200 m shown in green colour. This depth range values formed the highest number of depth solutions across the study area. The shallower depth solution values in blues (less than 150 m) mostly occur along linearized bodies interpreted as geological dykes which occur closer to the surface. The derived deeper depth solution values (>300 m) in red occupy the southeastern part of the study area, where the metavolcanics occur. From the depth estimates, it can be concluded that the overburden rocks in the study area is approximately between 150 m - 300 m thicker in most parts of the study area except the southeast, where much thicker sediments occur along the shoreline. The Advance Source Parameter Imaging Method measured relative depth solutions values to the overburden thickness at different location within the study area (Table 1). The measured depth solution values at various location is given alongside their geographical coordinates in X and Y. The highest estimated depth solution value in the study area is 753 m at the geographical coordinates X = 768700 and Y = 592700. The measured depth solution values are important for establishing mine drilling parameters, geohazard analysis study and are considered during mine development planning.

Figure 8.Advanced source parameter imaging depth solution map of the study area.

Table 1.Estimated depth solution values of the study area.