Application of remote sensing in mapping hydrothermally altered zones in a highly vegetative area - A case study of Lolgorien, Narok County, Kenya

Objective: To map areas of possible hydrothermal alteration using remote sensing technology; To map geological structures controlling mineralisation using remote sensing; To carryout ﬁeld mapping to ground truth the features identiﬁed by remote sensed data. Methods: Landsat 8 Operational Land Imager (OLI) and Shuttle Radar Topography Mission (SRTM) remote sensed images were downloaded from USGS website. Landsat images were processed using band ratios, band composites, principal component analysis techniques in ArcGIS to map areas of possible hydrothermal alteration. SRTM image was analysed using hillshade analysis technique in ArcGIS to map geological structural features controlling mineralisation. Findings: The study found that in several areas especially Southern, South Eastern, Central and North Western part of Lolgorien, there is a possibility of hydrothermal alteration as spectral signatures associated with iron oxide and clay minerals were identiﬁed. It was also found that areas with possibility of hydrothermal alteration are also associated with relatively large number of lineaments. It was also found that Southern and South-Eastern part of Lolgorien are also associated with numerous artisanal mines proving the fact that indeed gold mineralisation may be found in these places. However, due to thick vegetation cover, mapping of diﬀerent types of lithological units found in Lolgorien was not possible. Novelty/Application: The application of remote sensed technology helped in identiﬁcation of new areas of possible mineralisation such as Central and North-Western parts of Lolgorien which despite having similar geological properties (in terms of lineament density and hydrothermal alteration) have never been exploited for gold or other minerals.


Introduction
Even though it is not possible to directly see or delineate gold mineralisation using remote sensed satellite images, the mineralisation can be indirectly mapped using remote sensing as it is associated with several minerals which exhibit unique spectral properties (spectral reflectance) (1) . These minerals are usually found in hydrothermally altered zones. According to (2,3) , examples of these minerals are Kalinite, Illite (1M and 2M) and Dioctahedral smectite. The minerals have unique spectral signatures in the shortwave infrared section of the electromagnetic spectrum making them detectable by remote sensing (3) . As result, the spectral signatures associated with them can be used to map areas with possible gold mineralisation.
Generally, clay minerals (Kalinite, Illite and Dioctahedral smectite) are associated with gold mineralisation and, as a result, their identification may help in the search for gold mineralisation. Gold mineralisation may also be associated with iron oxides such as banded iron formations as is the case of Lolgorien and other places around the world. This means that delineation of banded iron formation and other iron oxides may lead to the finding of a gold mineralisation. Clay minerals and iron oxides are products of hydrothermal alteration and it is the hydrothermal fluids (mineralising fluid) that carry with them gold and other minerals. Therefore, areas that have undergone hydrothermal alteration may be mapped by identification of zones with clay minerals and iron oxides. This can be done using remote sensing.
The use of remote sensing in delineation of hydrothermal alteration (identification of clay minerals and iron oxides) as a way of mapping zones with potential of gold mineralization have been applied in several previous studies. A study by (4) utilised remote sensing approach to delineate hydrothermal zones and structures in Guelma basin, Northeastern Algeria. In (3) , integration of remote sensing technique with other techniques were used to map clay minerals (hydrothermal alteration) that maybe associated with gold mineralisatiomn in Roodepoort and Westonaria in Gauteng Province, and Witbank and Kriel in Mpumalanga Province, all of which are located in South Africa. They concluded that when interated with other techniques, remote sensing through mapping of clay minerals can be used delineate areas with possible gold (Au), Silver (Ag) and Tin (Sn) and Tungsten (W) mineralisation. In (5) , remote sensing was used to successfully delineate areas with iron bearing minerals (especially ferric and ferrous oxides) in Wadi Allaqi area, South Eastern Desert of Egypt. In (6) , remote sensing techniques were applied to discriminate areas with potential of iron-based minerals in Sar Cheshmeh copper mining district, south-eastern Islamic Republic of Iran using Landsat 8 (OLI) images. In this study, the band ratios with bright pixels helped delineated areas with potential of iron bearing minerals. (7) also applied remote sensing techniques such as band combinations, band ratios and composite to successfully map hydrothermal minerals associated with gold mineralisation in Southern Tianshan area, China. The researchers used ASTER data in this process, and they concluded that VNIR (visible and near-infrared) and SWIR (Shortwave infrared) bands were the most effective for mapping hydrothermal minerals from the ASTER (Advanced Spaceborne Thermal Emission and Reflection Radiometer) images. In (8)(9)(10) , remote sensing technique was used to accurately map gold mineralisation in Gauteng and Mpumalanga Provinces, South Africa by mapping clay minerals which are often associated with gold mineralisation.
In (11,12) , ASTER remote sensing imagery was used to map and discriminate phyllosilicate mineral groups and listvenite occurrences, in Antarctic environment of Northern Victoria Land. Using VNIR, SWIR and thermal-infrared (TIR) bands of ASTER, the authors were able to delineate Al, Fe3+-rich, Fe2+ and Mg phyllosilicates rich rocks. In (13) by using VNIR and SWIR bands of Landsat-8, Sentinel-2, ASTER and WorldView-3 (WV-3) remote sensed images, the authors were able to discriminate areas associated with Zn-Pb mineralisation in the central part of the Kashmar-Kerman Tectonic Zone, Central Iranian Terrane. In (14) , the authors used band ratio Principle component analysis to process Landsat Enhanced Thematic Mapper + (Landsat-7 ETM+), Landsat-8 and ASTER images so as to map hydrothermal alteration zones that may be associated with epithermal gold mineralization in the Ahar-Arasbaran region, North West Iran. In (15) , Landsat-8, ASTER and WorldView-3 multispectral remote sensing data were used to map hydrothermal alteration associated with copper-gold mineralisation in Northeastern Inglefield Mobile Belt (IMB), Northwest Greenland. In (16) , authors used Sentinel-1, ASTER, Phased Array type L-band Synthetic Aperture Radar (PALSAR) and Sentinel-2 to map regional structural units that control gold mineralisation in Barramiya-Mueilha Sector, Egypt. In (17) , multispectral and Synthetic Aperture Radar (SAR), Landsat 8 (OLI), ASTER, PALSAR and Sentinel-1B data were processed using band rationing, independent component analysis, principal component analysis, automated and semi-automated lineament extraction, and directional filtering to map zones of possible gold mineralisation and structural features controlling the mineralisation in South Eastern Desert, Egypt.
Lolgorien is one of the most active gold mining areas in Kenya hosting Kilimapesa Gold Mine and a number of artisanal mines. Despite its contribution to Kenya's gold mining industry, it is one of the most understudied geological settings in Kenya.
Additionally, despite the current improvement in mineral exploration technologies such as application of remote sensing technique, little effort has been made to update the mineralised zones, structural features and geological units using the technique. This means that gold mineralisation and the associated minerals (such as iron oxides and clay minerals) in the area have never been mapped using remote sensing technique. Furthermore, no recent geological study has been carried out https://www.indjst.org/ in the area except for the ones carried out by interested mining companies. The most recent known (to the best knowledge of the authors) geological study carried out in the area other than the ones being carried out by mining companies was carried out in 1991 by (18) . But (18) 's study never incorporated remote sensing in identification mineralised areas. A study by (19) in the 1940s which was done on behalf of the Geological Survey of Kenya, mapped geological units found in the study area. It is, however, important to note that by this time (1940s), remote sensed data were not readily available and it is highly unlikely that Shackleton (1946) used remote sensed images to map Lolgorien's gold mineralisation and geological units. This study attempts to bridge this gap by employing a relatively modern technology (remote sensing technology) to map hydrothermal minerals such as clay minerals and iron oxides which may be associated with gold mineralisation in Lolgorien.

Study location
Lolgorien Sub-county is located in Western part of Kenya within the Great Rift Valley, in Narok County. It is approximately 223km from Nairobi (the Kenyan capital city), and its approximate geographical coordinate is between 1 • 09' 0" S to 1 • 15' 0 S and 34 • 44' 0" E to 34 • 55' 0" E. Lolgorien area receives moderate rainfall and as a result, the area is covered by moderate to thick forest (vegetation) is some areas and grass lands in other areas. The location of the study area is shown in Figure 1.

Geological setting
Lolgorien is located within the Archean geological system of the Western Kenya. This geological system often consists of Nyanzian and Kavirondian units which are part of the larger Tanzanian craton and are considered to be the oldest rocks in Kenya (20) . Kavirondian systems which rest on top of Nyanzian system mainly consist of greywackes, grits, conglomorates and sandstones. The Nyanzian system is mainly composed of lavas, pyroclastics, banded iron formations as well as minor sediments. It mainly consists of the basalts, cherts, iron stones (banded iron formation) and shales (20) . These systems (Nyanzian and Kavirondian) are isoclinally folded and their axes generally trend in the East-West direction (20) . These two rock systems are generally intruded by several granitic and batholith intrusions. It is important to note that Tanzanian craton which hosts Nyanzian and Kavirodian systems is generally associated with precious and base metal mineralisation (20) . As such, metallic minerals such as copper, gold and silver have been reported in the area (20) . The geology map of lolgorien is shown in Figure 2.  Each of the two packages consisted of 11 bands images in form of GeoTIFF files, 1 metadata file (in form of ASCII file) and 1 quality assessment band image in form of GeoTIFF image. The projection of the downloaded Landsat 8 images is UTM zone 36S and the datum is WGS 84. These images were downloaded via the Earth Explorer which is an interface that allows access to the United States Geological Society (USGS) which is remote sensing database where Landsat and other remote sensed data are archived. Of the 11 bands included in each of the downloaded package, the visible, NIR and SWIR bands were of particular interest. Specifically, the following bands were of particular interest in this study: band 2 to 7. These are the most appropriate Landsat 8 bands for mineral exploration mapping, according to (21) .

Data collection
Other than Landsat images, Shuttle Radar Topography Mission (SRTM) Digital Elevation Model (DEM) data with resolution of 30 m was also used in the study. It was used in mapping of the geological lineaments that may be found in the study area. SRTM DEM image had an identification of number (ID) as SRTM1S02E034V3 and a resolution of 1-ARC second (30 m). It was acquired on 27th August, 2020 from United States Geological Survey (USGS) website.

Data processing
In order to ensure that geolocal features captured by the satellites are enhanced and can be correctly identified, the downloaded images were processed using the following enhancement techniques: 1. Band rationing 2. Band compositing 3. Principal Component Analysis.

Band compositing (band combination)
The following band combinations were used: 4,2,3 (True colour image); 5,4,3 (False colour image); 7,5,2 and 5,6,7. These band composites are shown in Figure 3. When composite image is created using a combination of 3 bands (such as infrared bands or infrared bands and visible bands) in RGB, an image that is produced tends to enhance certain features (22) . This, however, depends on the bands selected for compositing. These bands are often assigned using the rocks' or altered minerals' spectral characteristics. The band combination for identification of altered minerals, according to (6,21) , are: RGB (5,6,7), RGB (7,5,2) and RGB (5,4,3). It is for this reason that these band combinations were used to discriminate hydrothermally altered zones within the study area. The results for band combination are shown in Figure 3.

Band rationing
Band rationing is satellite image transformation technique where digital value (also known as the brightness values) of one particular band is divided by that of another band (23) . This technique of band combination, according to (23) , enhances as well as improves the compositional information of a satellite image while at the same time suppressing information that may not be useful. An example of such information (suppressed information) is shadowing caused by the topography. This allows for identification of the features that may not be seen in the unprocessed data. There exists several band ratios and false colour composites associated with these band ratios that have been proposed for enhancing alteration zones and lithological features. For instance, for Landsat 7 ETM+, band ratio 3/1 was suggested for discriminating iron oxides, and band ratio 5/7 has been suggested for discriminating hydroxyl-bearing minerals. For landsat 8 OLI, band ratios and band combination of 4/2, 6/7, 6/5 and 7/5 in RGB are essential in discriminating lithological units, altered rocks and vegetation (21) .
https://www.indjst.org/ As such the following band ratios were used in this exercise 1. Band ratio 4/2 was used to highlight iron oxide. 2. Band ratio 6/5 was used to highlight ferrous minerals (iron bearing) other than iron oxides such as amphiboles, olivine and pyroxyne. This is because these minerals tend to have high adsorption in band 6 and reflectance on band 5. 3. Band ratio 6/7 was used to highlight hydroxyl bearing rocks. 4. Band ratio 7/5 was used to highlight clay minerals.
The results of band ratios are shown in Figure 4.
In addition to band ratios, band ratio composites were also formed. The following band ratio composites were formed: Sabin's ratio (4/2, 6/7 and 6/5 in RGB) which was used to map hydrothermal alteration and in mapping of the lithological features; Kaufmann's ratio (7/5; 5/4; 6/7) which was used to discriminate altered rocks and lithological units from the vegetation; and composite band ratio 4/2, 6/7, 5 in RGB which was also used to discriminate altered rocks and outcrops from vegetation. The results of band ratio composites are shown are in Figure 5.

Principle component analysis
The density of the vegetation plays a major role in the detection as well as mapping of the hydrothermally altered rocks using band rationing. In order to minimise the effects of vegetation, a spectral unmixing technique known as principal component analysis was employed in this study (24) . Process of principal component analysis was begun by carrying out the component principle analysis on six landsat 8 bands, namely: band 2, band 3, band 4, band 5, band 6 and band 7. The results of this analyis are shown in Tables 1 and 2 (eigen vector matrix output) and Figure 6 A and B. The results in Tables 1 and 2 identify the principal components (PCs) that contain the most useful spectral information.

Hillshade analysis
Shuttle Radar Topography Mission (SRTM) Digital Elevation Model (DEM) image was analysed using hillshade technique (315 o azimuth and 45 o dip) to discriminate the lineaments from other relief features. From the hillshade image, lineaments were extracted manually and results are shown in Figure 7A and B.

Band combination
The results of the various band combinations are shown in Figure 3. The Figure 3A is true colour image created by combining bands 4, 3 and 2 in RGB. Green colour represents vegetation and dark brown colour represents exposed areas. Whitish colour represents buildings. Figure 3B is a false colour image created by combining bands 5, 4 and 3 in RGB. Brown and grey colours represent outcrops (exposed areas), whitish colour represents areas with houses (areas occupied by people), and red colour represents vegetation. Figure 3C is a composite of bands 7, 5 and 2. Orange (Brown) colours represent outcrops, black colour represents water (river) and green colour represents vegetation. Figure 3D is a composite of Band 5, band 6 and band 7. Light blue colour represents exposed surfaces, black colour represents water bodies, hydrothermally altered rocks were represented by blue colour, and vegetation is represented by brown and orange colours. https://www.indjst.org/

Band ratio
The results of band ratio are shown in Figure 4. The Figure 4A is band ratio 4/2. This band ratio does not clearly discriminated areas of interest. Figure 4B is band ratio 6/5; Figure 4C is band ratio 6/7; and Figure 4D is band ratio 7/5. The bright tones (whitish regions) in B and D, and darker tones in C represent areas with potential of having minerals associated with hydrothermal alteration such as clay minerals such as montmorillonite, kaolinite and illite, and iron oxides. https://www.indjst.org/

Band ratio composite (Sabin's ratio and Kaufmann's ratio)
The results of band ratio composite are shown in Figure 5. The results show that Sabin's ratio ( Figure 5A) and 4/2, 6/7, 5 composite band ratios ( Figure 5C) were able to discriminate areas associated with outcrops from the vegetation. In Sabin's ratio ( Figure 5A), outcrops (hydrothermally altered outcrops) are shown in blue, vegetation is highlighted in green and water bodies in black. In composite of 4/2, 6/7, 5 in RGB ( Figure 5C), the areas associated with hydrothermally altered rock outcrops are shown in reddish brown colour, vegetation is shown in green colour (both light and dark), and water bodies by black pixels. Kaufmann's ratio ( Figure 5B) showed that these outcrops are probably associated with hydrothermally alternated units. In this https://www.indjst.org/ composite (7/5, 5/4, 6/4 in RGB) and as shown in Figure 5B), rose colour (violet) represent areas that are probably associated with outcrops associated with granite or basalt as these rocks are often associated with amphibole, olivine and pyroxene; black represents water bodies (in this case rivers), red represents areas probably associated with iron oxides, and light blue represent vegetation.

Principal component analysis
To highlight or discriminate hydrothermally altered rocks and other minerals from vegetation, bands 1, 2, 3, 4, 5 and 6 were processed using principal component analysis. The Eigen values and Eigen vectors associated with this analysis are shown in Table 1. From the Table 1 it is observed that PC1, PC2 and PC3 contain more than 99.85% of the data variance. They were, therefore, combined in an RGB composite to yield principal component image shown in Figure 6A. From the figure, the green colour in the figure represents vegetation; and blue colour represents hydrothermal alteration rocks.  Table 2. The results are shown in Figure 6B. From the Table 2 it is observed that PC4 contains 0.09% of the total variance data, and it shows the highest positive Eigen vector value loading for band 2 (0.73372) and the highest negative Eigen vector value in band 4 (-0.65951). Generally, minerals associated with iron oxides have low reflectance ranging between 0.45-0.51 µm (range of band 2) and low absorption ranging between 0.64-0.67 µm (band 4 range). Therefore, in these two bands (band 2 and band 4) pixels associated with iron oxides are often bright in the PC4 image as shown in Figure 6B.

Lineament extraction
Lineaments in the study were extracted using hillshade analysis of the SRTM-DEM image of the study area. The extracted lineaments are shown in Figure 7A and B. Figure 7A is the lineaments plus hillshade map of the area of the study. Figure 7 B is lineament density map. Figure 7C is the rose diagram of the identified lineaments. From the figures it is observed that there is higher density of lineaments in the southern and South Eastern part of Lolgorien which happen to the same areas associated with iron oxides and other hydrothermally altered rocks. Rose diagram ( Figure 7C) show that most of the lineaments trend in the NW-SE direction which is the same trend as the lithological units found in the study area. https://www.indjst.org/

Discussion
Lolgorien area is generally covered by vegetation (some areas with thick vegetation and others with light vegetation) making it difficult to map geological features and contact zones (geological boundaries) associated with lithological units using remote sensed images especially the Landsat images (Landsat 8). Examples of regions covered by thick forests and light vegetation are shown in Figure 8 A and B respectively. Even though the area is covered by both light and thick vegetation which make it difficult to map geological boundaries using Landsat images, the remote sensing exercise was successfully used to map lineaments and areas associated with hydrothermal alterations (especially the clay minerals and iron oxides).

Band ratio compositing in mapping of outcrops and iron oxides
A number of techniques were used to successfully to map outcrops and hydrothermal alteration zones especially those associated with iron oxides, ferrous minerals (minerals associated with iron ions such as amphibole, olivine and pyroxene other than iron oxides) and clay minerals. One of the techniques used as already discussed in the results sections is composite band rationing (Sabin's ratio, Kaufmann ratio and 4/2, 6/7, 5 in RGB band composite). The results of these composite band ratios are similar. They are shown in Figure 5. As shown in Figure 5, Sabin's ratio ( Figure 5A), hydrothermally altered outcrops are shown in blue, vegetation is highlighted in green and water bodies in black. In composite of 4/2, 6/7, 5 ( Figure 5 C), hydrothermally altered rock outcrops are shown in reddish brown, vegetation is shown in green colour and water bodies by black pixels. In Kaufmann's ratio (Figure 5 B), rose colour (violet) represent areas that are probably associated with outcrops associated with granite or basalt as these rocks are often associated with amphibole, olivine and pyroxene; black represents water bodies (in this case rivers), red represents areas probably associated with iron oxides, and light blue represent vegetation.
A field observation (ground truthing) and geological map confirmed the results of these composite. For instance, the lower part (southern) part of the study area is dominated by granite outcrops (see geological map of Lolorien in Figure 2). The areas around the Lolgorien hill and the South Eastern part of the study area are associated with banded iron formations (which are iron oxides) enclosed in basaltic formations. Figure 9A shows an example of granite outcrops found in the southern part of the study area while Figure 9B as an example of banded iron formations rocks found in the areas around the Lolgorien hill.
The technique of discriminating outcrops from the vegetation using band composites has been employed by other studies too. In (21) , band ratio composites were used to discriminate outcrops in Chaves area (Portugal) using Landsat 8 images.

Mapping clay minerals and iron oxides using band ratios
Band ratios were able discriminate areas associated iron oxides, ferrous minerals (other than iron oxide such as amphibole, olivine and pyroxene) and clay minerals. In the band ratios 4/2, 6/7, 6/5, and 7/5 as shown in Figure 4A, B, C and D respectively, the brighter pixels in Figure 4B and D and darker tones in C represent areas with possibility of having minerals associated with hydrothermal alteration such as clay minerals (montmorillonite, kaolinite and illite) and iron oxides. Band ratio 6/5 highlighted areas with abundant ferrous minerals (other than iron oxides) in bright pixels as shown in Figure 4C. The band 7/5 discriminated areas with potential of rocks associated with clay minerals such as kaolinite, illite and montmorillonite as shown in Figure 4D. Band ratio 6/7 discriminated areas that possibly have altered rocks associated with alunite and clays ( Figure 4B). The ground truthing exercise (field observations) proved this as shown in Figure 9B. The figure is an outcrop of banded iron formation (which is associated with iron oxides) found around Lolgorien hill.
The band ratio composite approach has been used in the past to discriminate hydrothermally altered rocks in heavily vegetated area. One such study was conducted by (25) which was conducted in Baguio district in Philippines using Landsat Thematic Mapper (TM) images as well as ground mapping. The study found that the brighter pixels found in band ratio analyses were associated with hydrothermal alterations. Another study that employed band ratios 4/2, 6/7 and 6/5 in mapping hydrothermally altered rocks was the study by (26) which was carried out on Landsat 8 images in Ariab Mining District, Red Sea hills, Sudan. In (26) , the brighter pixels associated with these band ratios discriminated areas of possible hydrothermal alterations such as areas of iron oxides, ferrous minerals and clay minerals. Authors in (5) also used band ratios to successfully delineate areas with iron bearing minerals (especially ferric and ferrous oxides) in Wadi Allaqi area, South Eastern Desert of Egypt. In (6) , band ratios were used to discriminate areas with potential of iron-based minerals in Sar Cheshmeh copper mining district, south-eastern Islamic Republic of Iran using Landsat 8 (OLI) images. In this study, the band ratios with bright pixels delineated areas with potential of iron bearing minerals.

Mapping of outcrops using band combination
The band combinations were mainly used to discriminate areas covered by the outcrops from the those covered by vegetation and water. Band combinations: 7,5,2; 5,6,7; 5,4,3; and 4,3,2 were used in this case. The results are shown in Figure 3 (A to D). In all these 4 band combinations, rock outcrops and exposed surfaces were successfully discriminated from the vegetation and water bodies (rivers in this case). The use of band combination to discriminate rock outcrops has been employed in the previous studies. One such study was conducted by (21) who used the above band combinations to discriminate rock outcrops from thick vegetation in Chaves license, Portugal. A similar study was carried out by (27) who used band combinations on Landsat 8 imagery to delineate rock outcrops from clouds, sea and snow in Antarctica. In (28) , band combinations were also used on Landsat 8 OLI data (the same band combinations as the ones used in this study) to successfully delineate hydrothermally altered rocks and rock outcrops at Singhbhum Shear Zone in East India.

Mapping of hydrothermal alteration zones using principal component analysis
To further delineate areas associated with iron oxides and hydrothermal alteration zones, principal component analysis was carried. This analysis was applied because it minimizes the effect of vegetation. The results are shown in Tables 1 and 2 and Figure 6. As shown in Figure 6A and B, the hydrothermal alteration zones are shown by bright pixels in Figure 6B and blue pixels in Figure 6A. The use of principal component analysis in identification of areas that may associated with hydrothermal alteration and iron oxides has been carried out before. A study by (21) used this technique to successfully delineate hydrothermal zones in Chaves license, Portugal.

Relationship between hydrothermal alteration and lineaments occurrence
From the lineament maps shown in Figure 7 A and B, it is observed that areas identified as possible hosts of minerals associated with hydrothermal alteration have large number of lineaments. This is because lineaments (especially faults) are potential pathways through which mineralised fluids (hydrothermal fluids) migrate and alter or mineralise the adjacent wall rock or form mineralised veins. As such, areas with high density of lineaments are most likely associated with hydrothermally altered rocks. https://www.indjst.org/

Conclusion
Despite the study area covered by thick vegetation, processing of Landsat 8 (OLI) images using band ratio, band composite and principal component analysis, and processing of SRTM image using hillshade analysis were able to identify areas of possible hydrothermal alteration (areas that may be associated with clay minerals and iron oxides), and areas associated with high density of lineaments. These areas may be related to Lolgorien's gold mineralisation. The areas are mainly located in the Southern, South Eastern, Central and North Western part of the study area. These were confirmed by field work as exposed banded iron formations were found in a number of locations especially the Southern and South-Eastern part of Lolgorien. The study also found that areas with possibility of hydrothermal alteration (the identified areas) are also associated with relatively large number of faults. Coincidentally, these areas are also associated with numerous abandoned artisanal mines proving the fact that indeed gold mineralisation may be found in these places. It was also noted that even though Central and North-Western parts of Lolgorien had similar geological properties (in terms of lineament density and hydrothermal alteration) as the Southern and South-Eastern parts, these zones (Central and North-Western parts) have never been exploited for gold or other minerals. It is, therefore, recommended that future studies consider geochemical and geophysical characterisation of these zones ascertain the presence of these minerals. It is important to note that due to thick vegetation cover, mapping of different types of lithological units found in Lolgorien was not possible.