ISSN XXXX XXXX © 2018 IJESC

Research Article Volume 8 Issue No.1

Gold Potential Mapping using Remote Sensing and GIS at the

Prestea Concession of Golden Star Bogoso/Prestea Ltd, Ghana
T. Botwe1, E.M. Osei Jnr2
M.Sc. Student1, Senior Lecturer2
Pan African University Institute of Earth and Life Science Including Agriculture (PAULESI), University of Ibadan, Nigeria1
Kwame Nkrumah University of Science and Technology, Kumasi, Ghana2

Abstract:
Remote sensing and data integration techniques using Geographic Information Systems (GIS) offers a less expensive and
attractive strategy in mineral exploration, notably for identification of target areas in the screening phase of a new mineral
exploration project. Exploration companies in Ghana have not embraced and explored this powerful tool. The research aimed at
integrating remote sensing data and field data through a knowledge driven approach in a Geographic Information System (GIS)
environment to map areas of potential gold mineralization at the Prestea concession of Golden Star Bogoso/Prestea Ltd, in the
Prestea Huni Valley District of the Western Region of Ghana. Landsat 7 ETM + data were processed through band ratioing to
highlight the hydrothermal alterations relevant to gold mineralization within the study area, as well as extraction of lineaments
using the automatic extraction method. The alterations and lineaments, together with geological structures mapped during field
mapping, soil geochemical data, and the lithologies were reclassified into a common suitability scale and the results integrated
through weight overlay in a model built using ArcGIS’s model builder tool to produce the gold potential map of the study area. A
total of 995 known gold deposits were used for validation. The gold potential map outlined an area of 39.40 km 2 which
represented 33.94% of the total area as having high potential or favorability for gold mineralisation. The results of the study
indicate that the best predictor for gold mineralisation within the study area is geological structures followed by hydrothermal
alterations.

Keywords: Ghana,GIS,Gold Potential Mapping, Prestea, Remote Sensing

I.INTRODUCTION analysis through GIS in a mineral exploration context to

identify gold-rich potential areas in the Birim North District of
A company that implements mineral prospection or the Eastern Region of Ghana and established that the best
exploration programs usually carries out work at many predicators were the geochemical, geophysical factors and
different scales (from small to large) with the objective to lineament, while alteration is the least predictor. The gold
release barren and to retain potentially mineralized areas, out potential map delineated 158 km2 as gold potential area,
of initially acquired exploration areas, until areas worthy representing 32% of the entire area of 497 km2 as favorable for
enough for drilling are discovered. During these mineral the occurrences of gold deposits. Chandrasekar et al.,[2] used
exploration programs, abundant data of different types is multispectral satellite data to map out the heavy mineral
generated. Various datasets from geological, geochemical, deposits in South Tamil Nadu Coast of India. Minerals which
geophysical surveys, different remote sensing (RS) data show much variation in reflectance at different spectral bands
(including satellite images and aerial photographs) topographic mapped out. This has opened up new areas for inland heavy
data, have to be managed and integrated to define target areas. mineral exploitation and leads to eco-friendly mining of
In the past, such integration of multiple datasets was done via natural resources along the study area. It also showed the high
analogical methods, overlapping different layers of potential of multispectral satellite data for exploration and
information by means of transparencies in an attempt to mapping of mineral resources. Xianfeng et al.,[3] in their work
discover relationship between layers. Today this integration of mapped lithologies related to gold deposit in the South
datasets can be done digitally using a Geographic Information Chocolate Mountains area, California using Advanced
System (GIS).The use of GIS and remote sensing in mineral Spaceborne Thermal Emission and Reflection Radiometer
exploration has the following advantages over the conventional (ASTER) radiance data and then compared different methods
method: Exploration works can cover large areas at a reduced for extracting mineralogical information from ASTER data .
cost and a fast rate when using remote sensing and GIS as Andrada-de-Palomera,[4] used Landsat ETM+ and ASTER
compared to the conventional method. Various exploration image to delineate the potential gold areas in the Deseado
datasets from geological, geochemical, geophysical survey as Massif, Southern Argentina; followed by extracting the
well as data from remotely sensed images can be combined mineralogical indicative features from the ASTER data and the
readily when using remote sensing and GIS as compared to the indicative hydrothermal alteration from Landsat data. These
conventional method. data were combined with other geologic data like lineaments.
It was apprehended that, the weight of evidence approach was
A. Previous Work on Mineral Exploration successful to produce mineral predictive maps necessary to
Various authors have researched into the usage of GIS and investigate the spatial association of Low Sulfidation
Remote Sensing in the field of mineral exploration. Kwang et Epithermal Deposit (LSED) with lithology, structures and
al.,[1] used remote sensing methods and geographic data hydrothermal alterations as well as to deduce some information

International Journal of Engineering Science and Computing, January 2018 15891 http://ijesc.org/
related to the genesis of LSED in the Deseado Massif. He also western part has lower hills generally ranging in height
concluded that Simple band ratios (5/7 for Landsat and 4/6 for between 70 m and 110 m.
ASTER) proved to be more suitable methods than principal
component analytical techniques to map clay alteration at
regional to district scale using Landsat and ASTER data.
According to him, a great advantage of using ASTER over
Landsat data is the ability to map silica abundance, which is an
important indicator for LSED in the Deseado Massif.

B. Knowledge Driven Approach

Knowledge-driven approach in mineral potential mapping is a
more subjective approach and relies on the modeler’s input, or
expert opinion or knowledge concerning mineralisation to
provide the relative weightings assigned to the different
evidential themes. These methods may be more appropriate for
lesser to unexplored landscapes (Greenfield areas) where there
are few known deposits of the type sought [5].Examples of
knowledge-driven model types include; Fuzzy Logic, Boolean
overlay and Weight overlay.

C. Weight Overlay
A group of methodologies applied in optimal site selection or
suitability modeling is referred to as overlay analysis. It is an
approach for assigning a common scale of values to varied and
dissimilar inputs to create an integrated analysis. Suitability Figure.1. Location of the study area
models define the best or most preferred locations for a
specific phenomenon. Overlay analysis most often entails the B. Geology of the study Area
examination of many different factors. For instance, selecting a The Prestea concession lies within the southern portion of the
site suitable for gold mineralisation means assessing Ashanti Greenstone Belt along the western margin of the belt.
parameters relevant to gold mineralisation within the locality Rock assemblages from the southern area of the Ashanti Belt
such parameters may include lithology, hydrothermal were formed between a period spanning from 2,080 to 2,240
alterations, and geological structures among others. These data Ma, with the Sefwi Group being the oldest rock package and
exist in different rasters with different value scales. A raster of the Tarkwa sediments being the youngest. The Ashanti Belt is
hydrothermal alterations (DNs) cannot be added to a raster of host to numerous gold occurrences, which are believed to be
distance to geological structures (meters) and obtain a related to various stages of the Eoeburnean and Eburnean
meaningful result. More so, the factors in the analysis may not deformational events. The geology of the Prestea concession is
have equal importance. It may be that hydrothermal alteration divided into three main litho-structural assemblages, which are
has good correlation to gold mineralisation than the distance to fault bounded and steeply dipping to the west. This suggests
geological structures. The importance of these parameters are that the contacts are structurally controlled and that the litho-
determined by the modeler and shown as weights which sum structural assemblages are unconformable. From the eastern
up to 100% or 1. footwall to the western hanging wall, these packages are
represented by the Tarkwaian litho-structural assemblage, the
II. MATERIALS AND METHODS tectonic breccia assemblage, composed of sheared graphitic
sediments and volcanic flows, and the last assemblage is
A. Study Area composed of undeformed sedimentary units of the Kumasi
The Prestea concession is cited in the Western Region of Basin, which is located to the west of the Ashanti fault zone.
Ghana about 200 km from the capital Accra and 50 Km from The Tarkwaian litho-structural assemblage to the east is mostly
the coast of the Gulf of Guinea with latitude 5.4373 0 N and composed of sandstone, pebbly sandstone, and narrow
longitude 2.14010 W. Bogoso and Prestea comprise of a conglomerate units. Bedding and sedimentary textures have
collection of connecting mining concessions that together been observed sporadically, and in most cases they have been
cover a 40 km section of the Ashanti gold district in the central obliterated by hydrothermal alteration and deformation at the
eastern section of the Western Region of Ghana. Entry to the proximity of the Ashanti fault. The litho-structural assemblage
Prestea concession by road is a six-hour drive from Accra overlying the Tarkwaian sediments is a tectonic breccia
through the port city of Takoradi. The road linking Accra to bounded to the west by the Kumasi sedimentary basin. The
Prestea is paved. There are airports at Kumasi and Takoradi, tectonic breccia is a polygenic assemblage, composed of
providing daily services to the international airport situated at various rock types such as volcanic rocks, volcanoclastics,
Accra. Kumasi is situated about a three and half hour drive sediments of the Birimian Supergroup, and sparse Tarkwaian
from Prestea. Road surfaces in the area vary from poor (on the sedimentary slivers. Volcanic lenses have been divided into
section between Bogoso and Prestea) to good (Accra to two units based on their alteration pattern: weakly altered
Bogoso). The topography of the area within which the Prestea mafic volcanic rocks are characterized by a distal
concession is located generally slopes in a northern direction chlorite/calcite alteration pattern, while strongly altered mafic
towards the Ankobra River. It can be described as gently volcanic rocks are characterized by proximal silica/ sericite/Fe-
rolling, punctuated by a number of low hills and rises. Series Mg carbonates alteration pattern. These strongly altered mafic
of NE-SW trending sub-parallel ridges, approximately 2 km volcanic lenses are generally located at proximity to the Main
wide, dominates the eastern part of the Prestea concession. Reef Fault or bounded by second order footwall faults. The
These ridges range in height from 150 m to 195 m. The tectonic breccia assemblage is believed to have been the focal

International Journal of Engineering Science and Computing, January 2018 15892 http://ijesc.org/
point of the post thrusting Eburnean deformational events (syn- 2002. The image has WRS path of 196 and WRS row
D3 to syn-D5), therefore, primary textures, whether syn- of 56)
volcanic or syn-sedimentary, have only been locally preserved.
Volcanic lenses are intercalated with sheared graphitic E. Software
sedimentary horizons which represent strained and brecciated This study made use of ESRI ArcGIS version 10.4.1. The
sequences of siltstones, mudstones and greywacke units ESRI ArcGIS Spatial Analyst extensions were used to run
affected by pervasive graphitic alteration. Primary textures are most of the analysis (IDW interpolation, calculation of
generally overprinted and obliterated by deformation, but euclidean distances, reclassification among others). The core
bedding has locally been preserved. The most western litho- geological predictions were made using weight overlay tool
structural assemblage underlying the Prestea concession under the spatial analyst extension. The ENVI 4.7 software
consists of relatively unreformed to weakly strained was used in the image processing. PC1 Geomatica software
sedimentary rocks of the Kumasi basin. The assemblage is was also used in the extraction of lineaments from the satellite
composed of a series of flyschoid sequences where the most image.
common units found are argillites, mudstones, siltstones and
greywackes, which are all commonly referred to as phyllite in F. Method
Ghana. Several syn-sedimentary textures have been observed The flowchart of the method used in this research work is
such as bedding planes, graded bedding and cross-bedding. illustrated in Fig.3. Georeferencing was performed on the
Chert horizons are locally intercalated within the flysch geological map of the study area because of difference in
sequence, but appear to lack lateral continuity. The major coordinate system between the geological map and the image,
lithologies are illustrated in Figure 2. the map was then digitized to form the vector layer. Pre-image
processes such as stacking and subseting were performed on
the ETM+ image to obtain the imageof the study area. Some
image enhancement techniques such as band combination,
band ratioing, and principal component analysis were
performed on the image of the study area for the extraction of
hydrothermal alteration and lineaments, and this formed the
raster layer. The point soil geochemical data of the study area
was converted into a continuous surface representation using
the inverse distance weighting interpolation in ArcGIS 10.4.
Geological structural data obtained through field mapping was
also processed on ArcGIS 10.4 to obtain the proximity to
geological structures. The lineaments extracted were also
processed to obtain the distances to lineaments. The vector
layer, the raster layers, the geochemical data, and the distances
to geological structures were all reclassified into a common
suitability scale. By using the weight overlay tool, the
reclassified data were combined, with their respective weights
attached to them to produce the gold potential map of the study
area using the model builder in ArcGIS 10.4 as illustrated in
Fig.4. The gold potential map was validated with known gold
occurrence points of the type sought. A total of 995 known
gold occurrence points were used for validation.

Figure .2. Major rock types within the study area

C. Exploration Model of the Prestea Area

The exploration model (prediction criteria) for mesothermal
gold deposits within the study area includes; alterations,
lithology, and geological structures. These features serve as
guides or evidence to gold mineralization within the Prestea
concession. These features together with the gold in soil
geochemical data were used as evidential themes in the
prediction of areas favourable for gold mineralisation within
the Prestea concession of Golden Star.

D. Materials

The materials used in this research work are listed below.

1. Gold occurrence points used for validation
2. Geochemical data of the Prestea concession
3. Geological structural data of the Prestea concession
4. Geological map of the Prestea concession
5. Remote sensing image (Landsat 7 ETM +) of the
study area. (The image was acquired on January 15,
Figure .3. Flowchart of methodology

International Journal of Engineering Science and Computing, January 2018 15893 http://ijesc.org/
G. Weighting of Evidential Themes

The weights were determined after a thorough literature review

on gold mineralisation within the study area. Furthermore,
consultations with mine and exploration geologists with
Golden Star Bogoso/ Prestea Ltd. on exploring for gold in the
study area resulted in the weighting in Table 1.

Table.1. Summary of weights attached to the evidential

themes.

Evidence % of influence Weight

Structures 50 0.5

Hydrothermal 25 0.25
alterations
Lithology 10 0.1
Geochemistry 15 0.15

Figure.5. Landsat 7 ETM + image of the study area in RGB

combination

Figure.6. Alteration map of the Prestea concession

Figure 6, shows the combined alterations mapped within the
study area. The red colour denotes iron oxide alterations,
yellow colour indicates clay mineral alterations as well as
carbonate minerals and chlorite alterations, the green colour
shows the ferrous mineral alterations. From Figure 6, it can be
Figure.4. Model for deriving the gold potential map observed that clay mineral alterations are the abundant
alteration patterns within the study area. Also, most of the
III. RESULTS AND DISCUSSION alteration patterns associated with gold mineralisation are
highly concentrated at the NE portion of the study area
The results obtained from the various image processing and
geoprocessing techniques are presented in this sub-section. B. Results of the Lineament Extraction and Processing of
Geological Structures
A. Results of layer stacking and subseting Figure 7, shows the lineament map of the Prestea concession.
Lineaments are mappable linear or curvilinear feature of a
The Figure 5 shows the results of layer stacking and subseting surface whose parts align in a straight or slightly curving
of the Landsat ETM+ image to obtain the image of the study relationship. They may be an expression of faults, joints,
area. lithological contacts, fractures and others. From Figure6, the

International Journal of Engineering Science and Computing, January 2018 15894 http://ijesc.org/
linear feature striking in NE-SW and dipping to the NW may
reflect the main fault known as the Ashanti fault which occurs
at the contact between the met sedimentary and met volcanic
rocks and hosts numerous hydrothermal gold deposits.
Figure7, shows the straight line distances to the lineaments.
Areas closer to this feature have significant correlation with
gold mineralisation than areas farther away. Figure 8, shows
the spatial distribution of mapped geological structures within
the study area. The geological structures include quartz veins,
shear zones, faults and foliation. These structural features
represent zones of weakness favorable for trapping and
accumulation of hydrothermal fluids resulting in localization
of ore deposits

Figure.8.Spatial distribution of geological structures

mapped in the Prestea concession
Figure.9. illustrates the proximity to geological structures in
the Prestea concession. Distances closerto structural features
have good correlation with gold mineralisation than distances
farther away hence distances closer to structural feature are
favourable for gold mineralisation. This is because during the
deposition of hydrothermal fluids, some of the fluids move
through the host rock via diffusion or the pore spaces within
the rocks, making areas closer to point of deposition having
anomalous abundance of ore elements, with a logarithm
reduction away from the points of deposition.

Figure.6. Lineament map of the Prestea concession

Figure.9. Proximity to geological structures

C. Results of the Inverse Distance Interpolation of the

Geochemical Data
Figure 10, shows the results of the interpolation of the gold in
soil geochemical anomalies using the inverse distance
weighting tool under spatial analyst in ArcGIS. Inverse
distance weighting (IDW) interpolation explicitly makes the
Figure.7. Distances to lineaments assumption that thing that are close to one another are more

International Journal of Engineering Science and Computing, January 2018 15895 http://ijesc.org/
alike than those that are farther apart. To predict a value for
any unmeasured location, IDW uses the measured values
surrounding the prediction location. The measured values
closest to the prediction location have more influence on the
predicted value than those farther away. From Figure10, areas
denoted by red color zone have high concentration of gold,
followed by those areas denoted by yellow color whereas areas
denoted by green color have low gold concentration.
Geochemical data is very important in exploration because it’s
a direct method which measures the exact amount of the
mineral sought for.

Figure.11. Gold potential map of the Prestea concession

E. Validation of the Gold Potential Map

The gold potential map of the Prestea concession was validated
with 995 known gold occurrence points and the results is
illustrated in figure 12.From figure 12, it can be observed that
most of the validating points falls within area predicted to
havehigh potential for gold mineralisation shown on the map
with colours blue and yellow with few falling within the area
shown in green colour. The results of the validation confirms
the usefulness and effectiveness of applying remote sensing
and GIS in gold exploration.
Figure.10. Gold in soil concentrations of the Prestea
concession

D. Result of Weighting and Combining Dataset

Weights were assigned to various evidential themes for

prediction of areas with high potential for gold mineralisation.
The results is as shown in Fig. 11.From the gold potential map
shown in Figure 11, areas with low ranking values, 1,2,3 and 4,
have low potential for gold mineralization. These areas may
not be favourable for gold exploration and hence need no
further work. Areas with high ranking values, 6 and 5, are
areas which are mapped to have high potential for gold
mineralization. These areas are shown on the map with colours
blue and yellow respectively. The gold potential map
demarcates an area of 39.40 km2 out of the total area of 116.08
km2to be highly favourable for gold mineralization which
represents 33.94% of the total area of Prestea concession. The
gold potential map provides more spatial information on gold
occurrence within the study area. This would help the mine to
widen their scope of search within the concession to areas
where they have not explored at a reduced cost in an effective
and efficient way. This gold potential map would also enable
management of GSR to visualize the extent of gold occurrence
and plan the mine in relation to building of bungalows and
offices, plants installation, construction of haul roads among
others. Figure.12. Validation of the gold potential map

International Journal of Engineering Science and Computing, January 2018 15896 http://ijesc.org/
IV. CONCLUSION

Based on the results and analysis, a gold potential map of the

Prestea concession has been produced through the application
of Remote Sensing and GIS. This demonstrates the usefulness
and effectiveness of the application of Remote Sensing and
GIS in the exploration for gold. The study indicates that the
best predictor for gold mineralisation within the study area is
the geological structures followed by the hydrothermal
alterations. The gold potential map of the Prestea concession
outlines an area of 39.40 km2 which represents 33.94% of the
total area as highly favourable for gold mineralisation. The
phyllite rocks are highly deformed due to presence of
geological structures and may be very receptible for hosting
the gold mineralisation. The use of Landsat ETM+ as the
source of information for recognition parameters such as
hydrothermal alteration and lineaments was very successful.
Remote sensing and GIS are powerful exploratory tools which
can be applied in the initial stages of mineral exploration to
narrow our search to areas of potentials high mineralisation,
hence government agencies in charge of mineral exploration as
well as mining and exploration companies in Ghana should
embrace this technology since it is less expensive, effective
and efficient, environmental friendly and most of the resources
required are open source. Also, gold potential maps should be
produced for all the districts in Ghana to enable the
government use them as documents to attract more investors
into the exploration and mining industry. This technique
should be applied to the exploration of other minerals such as
iron, copper, uranium among others. Golden Star Exploration
should carry out more detailed exploration at places that are
predicted to have high potential for gold mineralisation but are
out of their current working zones.

V. REFERENCES

[1]. Kwang, C., Osei Jnr. E. M. and Duker, A. A. (2014),

Application of Remote Sensing and Geographic Information
Systems for Gold Potential Mapping in Birim North District of
Eastern Region of Ghana, International Journal of Remote
Sensing Applications Volume 4 ,9pp.

[2]. Chandrasekar, N., Sheik Mujabar, P. and Rajamanckam,

G. V. (2011), Investigation of heavy-mineral deposits using
multispectral satellite data, International Journal of Remote
Sensing, 32, 8641-8655.

[3]. Xiamen, Z., Micha, P., Nornan, D. (2007), Lithologic and

Mineral Information extraction for gold exploration using
Aster data in the South Chocolate Mountains (California).
ISRRS Journal of Photogrammetry and Remote Sensing, 62,
271-282.

[4]. Andrada-de Palomera R. P. (2004), Application of

Remote Sensing and GIS for Mineral Predictive Mapping,
Deseado Massif, Southern Argentina, MSc. Dissertation,
International Institute for Geo-information Science and Earth
Observation, Enschede, The Netherland, 30-45, unpublished.

[5]. Carranza, E. J. M. (2009), Geochemical Anomaly and

Mineral Prospectivity Mapping in GIS: Handbook of
Exploration and Environmental Geochemistry; Vol 11.
Elsevier, UK, 351 pp.

International Journal of Engineering Science and Computing, January 2018 15897 http://ijesc.org/