Lithological and Structural Controls On Gold Mineralizaton in Buhweju. Uganda
Lithological and Structural Controls On Gold Mineralizaton in Buhweju. Uganda
Lithological and Structural Controls On Gold Mineralizaton in Buhweju. Uganda
SUPERVISORS:
Dr. T. Woldai
Drs. B. J. de Smeth
Inter-Relationship between
Lithology and Structure and its
Control on Gold Mineralization in
Buhweju area, SW OF UGANDA
SUPERVISORS:
Dr. T. Woldai
Drs. B.J.de Smeth
i
ACKNOWLEDGEMENTS
I would like to thank my advisors Dr. T. Woldai and Drs. B. J. De Smeth for constructive advises and
effort they have made for fulfilment of this research. Special thanks go to Dr. Frank van Ruitenbeek, Dr.
John Carranza and Mr. Chris Hecker for their help to make me familiar with new techniques of mineral
exploration. I would like to thank Dr. Nigel Stack and Dr. Sally Barritt for they gave me permission to use
high resolution geophysical imagery for this research. Warmest thanks to local people around Buhweju for
their undesirable help during field work.
ii
TABLE OF CONTENTS
1. Introduction ...........................................................................................................................................................7
1.1. Research background ..................................................................................................................................................7
1.2. Research problem ........................................................................................................................................................7
1.3. Research Objective ......................................................................................................................................................8
1.4. Research Questions .....................................................................................................................................................8
1.5. Research Hypothesis ...................................................................................................................................................8
1.6. Overall research methodology .................................................................................................................................8
1.7. Available data sets .................................................................................................................................................... 10
1.8. Study Area.................................................................................................................................................................. 11
1.9. Justification for using various remote sensing data set ...................................................................................... 12
1.10. Thesis organization .................................................................................................................................................. 13
2. Previous Work.................................................................................................................................................... 14
2.1. Regional geology, Structure and Mineralization .................................................................................................. 14
2.2. Local geology, strucutre and mineralization ........................................................................................................ 16
3. Multispectral Image Processing and Interpretation ..................................................................................... 19
3.1. Introduction .............................................................................................................................................................. 19
3.2. Correction for Vegetation Effect .......................................................................................................................... 19
3.3. Univariate Analysis of ASTER and Landsat ETM+ data ................................................................................ 21
3.4. Principal component analysis ................................................................................................................................. 23
3.5. Summary of multispectral processing and interpretation .................................................................................. 26
4. Surface Lineament Interpretation ................................................................................................................... 28
4.1. Introduction .............................................................................................................................................................. 28
4.2. Summary .................................................................................................................................................................... 32
5. Processing and Interpretation of Gamma Ray Data .................................................................................... 34
5.1. Introduction .............................................................................................................................................................. 34
5.2. Data processing......................................................................................................................................................... 35
5.3. Image Enhancement and Interpretation .............................................................................................................. 37
5.4. Summary of Radiometric data interpretation ...................................................................................................... 44
6. Processing and Interpretation of Aeromagnetic Data ................................................................................. 45
6.1. Introduction .............................................................................................................................................................. 45
6.2. Data processing and Interpretation ....................................................................................................................... 45
6.3. Summary on aeromagnetic data interpretation ................................................................................................... 50
7. Data integration and Geological Map Compilation ..................................................................................... 51
7.1. Introduction .............................................................................................................................................................. 51
7.2. Lithological boundaries ........................................................................................................................................... 51
7.3. Compiled Lineament Interpretation ..................................................................................................................... 56
7.4. Mineralization ............................................................................................................................................................ 56
8. Conclusion and Recommendations ................................................................................................................ 61
8.1. Conclusion ................................................................................................................................................................. 61
8.2. Recommendations .................................................................................................................................................... 62
iii
LIST OF FIGURES
Figure 1-1: Flow chart of the research methodology .............................................................................................. 9
Figure 1-2: Display areas the different data sets used in this study. .................................................................... 11
Figure 1-3: Location of the study area ..................................................................................................................... 11
Figure 3-1: NDVI Images of the study area ........................................................................................................... 20
Figure 3-2: Subscence of Landsat ETM and ASTER color composite images in (RGB) ............................ 22
Figure 3-3:Subscence of OIF color composite in RGB order ............................................................................ 22
Figure 3-4: Subscens with interpreted lithologies .................................................................................................. 24
Figure 3-5:. Subscens with interpreted Lithologies ................................................................................................ 25
Figure 3-6: ASTER and Landsat interpretation map of Buhweju ....................................................................... 26
Figure 4-1: Subscenes of A) ASTER 721 in RGB order ...................................................................................... 29
Figure 4-2: Interpreted lineament map .................................................................................................................... 29
Figure 4-3: Color shaded relief images..................................................................................................................... 30
Figure 4-4 :SRTM DEM Subscens of the area. ...................................................................................................... 31
Figure 4-5: SRTM DEM Subscens of the area. ...................................................................................................... 31
Figure 4-6: Interpreted lineament map .................................................................................................................... 31
Figure 4-7: Final interpreted surface lineament map of the area ......................................................................... 32
Figure 4-8: Lineament density map of the area ...................................................................................................... 33
Figure 5-1: A) Element weathering and gamma ray response ............................................................................. 35
Figure 5-2: Box plot showing the distribution of radioelement per litholohic units ........................................ 36
Figure 5-3: Images of individual radioelement grids ............................................................................................. 38
Figure 5-4: Images radioelement ratio grids ............................................................................................................ 38
Figure 5-5: Various IHS composite images of gamma ray data........................................................................... 39
Figure 5-6: Different enhancements made on Ternary image . .......................................................................... 40
Figure 5-7: Previously mapped lithology ................................................................................................................. 42
Figure 5-8: Lithological interpretations on gamma ray data ................................................................................. 42
Figure 5-9:Interpreted lithologic units from radiometric data ........................................................................... 43
Figure 5-10: Interpreted lithological map. Numbers indicate the newly identified units ................................ 43
Figure 6-1: Enhanced Aeromagnetic images of the area .................................................................................... 47
Figure 6-2: Vertical derivative images calculated from RTP ............................................................................. 48
Figure 6-3: A) Regional litho-magnetic domains interpreted from RTP. .......................................................... 48
Figure 6-4: Interpreted structural map of the area.. ............................................................................................... 49
Figure 6-5: Magnetic lineament density map .......................................................................................................... 50
Figure 7-1: Subscens showing interpreted lithological units ................................................................................ 52
Figure 7-2: Subscenes showing interpreted lithological units .............................................................................. 53
Figure 7-3: Field photos of different rocks ............................................................................................................. 54
Figure 7-4: Improved geological map of Buhweju ................................................................................................ 55
Figure 7-5: Lineament map of Buhweju .................................................................................................................. 57
Figure 7-6: Field data and lithological interpretation on mineralization. ........................................................... 58
Figure 7-7: Distribution of gold occurrence areas.. ............................................................................................... 59
Figure 7-8: The Twangiza and Gieta trends............................................................................................................ 60
Figure 7-9: Structural corridors ................................................................................................................................. 60
iv
LIST OF TABLES
Table 1-1: Radiometric characteristics of ASTER and Landsat ETM data ...................................................... 10
Table 1-2: Stratigraphic units of the study area. After(Reece, 1961) .................................................................. 17
Table 3-1: Correlation coefficient of Landsat ETM+ data in the study area.................................................... 21
Table 3-2: Correlation coefficient of ASTER VNIR-SWIR data of the area ................................................... 21
Table 3-3: The first six highest OIF index ranking of ASTER and Landsat ETM+ bands........................... 23
Table 3-4: Eigen vector loadings of principal components ................................................................................. 24
Table 3-5: Eigen values of the principal component of ASTER SWIR data of the area................................ 24
Table 3-6: Eigen values of principal components of Landsat ETM data and their variance ......................... 25
Table 3-7: Result of image interpretation based on image characteristics. ........................................................ 27
Table 4-1: Soble filter in four main directions applied in this study .................................................................. 29
Table 5-1: Correlation between radioelements ...................................................................................................... 36
Table 5-2: Radiometric correlation of the newly identified units ....................................................................... 44
v
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
1. INTRODUCTION
1.1. Research background
Geological maps with their subsequent derivative are considered to be the most reliable geosciences
information having immense economic and societal value. A geological map provides information not
only about the distribution and thickness of individual rock units but also reveals relationship among
strata and structures which provide insight to many aspect example mineral potential. Acquiring geological
information has nowadays greatly assisted by remote sensing technology as lots of works have already
been documented.
Analysis and integration of multi disciplinary geoscientific information is becoming an advanced
technique to ameliorate the geosciences knowledge. The problems in traditional geological mapping
like limited exposed outcrop, extensive regolith and vegetation cover and subtle changes in geological
features caused mapping largely based on geological inference. In integrated approach however, the
information in each data set is analyzed separately and finally integrated to resolve the uncertainties and
geological inferences. In integrated geological mapping the individual data layer is brought with
different spectral and spatial resolutions that enable to extract more information and even to map
subtle variations in geological feature, when integrated they are particularly useful.
The deeply dissected mountainous ridges in the east and the western rift valley are typical topographic
features characterizing the study area, Buhweju. The geology of the area is presumed to be complex which
undergone through different sequences of geological events accompanied by folding, refolding,
recrystalization and metasomatism. The area is covered by various lithological units ranging from
Precambrian metamorphic complex which have been extensively weathered and eroded to Pleistocene rift
filling unconsolidated sediments and volcanics.
Buhweju is one of the gold and base metal producing areas in the country. Various compilation works on
the mineral resources of Uganda showed wide spread alluvial gold in Buhweju and its surroundings. Three
sulphide veins have also been identified at Kitaka, Kampono and Kanyambogo lead-zinc mines emplaced
in shcists and gneiss. In these mines coarse crystalline gold also occur in vughs lined by quartz crystals.
The aim of this research is therefore to see the relationship between lithology and structure of the area and
to assess control on mineralization
Analyzing, interpreting and compiling the information contained in various data sets can result
in producing necessary information related to lithological map, structural map and various thematic
information before and after ground truthing. The geographic information system (GIS) has played
major role in integrating and overlaying the various geosciences data and then producing maps which
could be taken to the field to assist field mapping and ground truthing. Reliable geoscientific information
in the form of geological map, structural map and mineral potential map could then be produced
which are very vital for exploration and development of mineral resources.
8
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
possible. During field work lithological descriptions and structural measurements were collected at about
140 observation points (Appendix 1). In addition various abandoned and active mine areas were surveyed
GPS location and geological descriptions were taken.
9
1.7. Available data sets
The ministry of Energy and mineral development of the republic of Uganda has conducted airborne
magnetic and radiometric surveys during survey date 07/12/06-31/05/07. The data was acquired and
processed by Fugro airborne surveys. Survey flight lines with a spacing of 200m were flown in NE-SW
direction and tie lines with a spacing of 2000m in NW-SE direction. The survey was conducted with mean
terrain clearance of 80m and with a sample interval of 0.1s and 1s for magnetic and radiometric data
respectively.
10
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
Figure 1-2: Display areas the different data sets used in this study.
11
Figure 1-3: Location map of the study area
12
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
SRTM data was carried out by E. A. Owusu et al. (2006) which greatly improved the geological map of
the Kibi area in Ghana. Integrated approach aided with field data discriminated lithological units with
limited spectral contrast and covered by vegetation(Schetselaar et al., 2000). It is well established in
many publications that lithology and structure play a determinate role in emplacement and
localization of mineralization. Aeromagnetic a n d remote sensing data integration has been
(Chernicoff et al., 2002 ) to identify crustal lineament control on magmatism and mineralization in
northwestern Argentina. An integration of airborne and optical data have been utilized to map the
regolith cover in Western Australia(Wilford et al., 1997).
13
2. PREVIOUS WORK
2.1. Regional geology, Structure and Mineralization
14
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
late Archean granitic rocks of Northern Uganda and by Ikingura et al. (1992) on granitic rocks sampled
from Zaire, Burundi and SW Uganda. These have led to a conclusion that at least three distinctive tectonic
events have been taken place in SW of Uganda which were consistent with granitic emplacement ages in
various parts of kibaran belt.
15
2.2. Local geology, strucutre and mineralization
Explanation of the geology of Buhweju (Reece, 1961) and related reports are the only existing geological
information available in the area and all of which form the foundation of this work . According to the
explanation, rocks of Buhweju are classified in to five groups. The Igara group represents a sedimentary
sequence which has been metamorphosed and granitized units including schist, quartzite and granitic
rocks. The Buhweju group consist a sequence of psammitic and pelitic rocks unconformably lying on the
Igara group. They have recrystallized the basal formation granitized contemporaneously with the Igara
group. Ibanda quartzite forms an isolated outcrop and its relationship with Igara and Buhweju groups is
not known. Doleritic dykes and quartz veins represent intrusive rocks which are younger than the
Buhweju but older than the Pleistocene rocks. The Pleistocene rocks are poorly consolidated and volcanic
occupying the rift valley. The various volcanic fields consisting of fine grained and highly calcareous tuffs
were well explained by (Reece, 1961). The Buhweju group is further classified in to five different lithologic
units including Lukiri mudstone, Isingiro conglomerate, Lubare quartzite, Kasyoha shales and Munyoni
quartzite in the order of from bottom to top in lithological succession. The geological succession and
description of each geologic units of the area has shown in (Table 1-2).
The general geological history and sequence of geological events of the area have been summarized as
follows: (1) deposition of the Igara group under shallow water marine condition followed by intrusion of
doleritic rocks. (2) Folding of the Igara group about north trending axes, accompanied by low grade
(greenschist) regional metamorphism. (3) Deposition of the Buhwehu group as deltaic sediment and later
as shallow water conditions. (4) Folding of parts of the Buhweju group along WNW trends with some
refolding of the Igara group accompanied by local felspathisations and metasomatism. (5) Intrusion of
quartz veins. (6) Formation of rift valley and intrusion of dolerites and quartz veins. (7) Deposition of
Pleistocene sediments.
16
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
The structure of the area was described by (Reece, 1961) from both field data and thin section analysis
especially for the Igara and Buhweju groups. The structure of Ibanda quartzite is not known but some
explanations suggested that it has affected by tectonic thickening which caused its size disproportional.
Rift valley and associated Pleistocene rocks didnt exhibit visible tectonic structures resulting most
structural explanations were entirely of topographical.
Foliations and lineation were explained as common structures apparent on schist, quartzite and gneiss of
the Igara group. Almost parallel NW foliation has been described on gneissic rocks marked by mineral
banding and preferred direction of feldspar porphyroblasts. Foliations were also described on schist with
weak planes marked by white mica flakes. Interpretations have showed that two foliations were affected
the Igara group rocks. The first dominant foliation was associated with early folding events trending
NNW and nearly isoclinals nature. The second foliation (slip cleavage), evidenced by drag folds and
fracture cleavage, was found superimposed on the older folds. Beddings, cleavages and linear structures
were described on various lithological units of the Buhweju group which led to an interpretation that the
Buhweju group has been folded in WNW trending axes.
Generally the Igara group has folded isoclinically about steeply dipping northerly trending axial planes
with a refolding about WNW trending axes in the south west. The Buhweju group has folded about
horizontal in WNW trending axes with intensity of folding increase to the south.
According to the various unpublished reports, first mining operations were carried out by African smelters
to extract iron ore from ferruginous brecciaed zones in the kasyoha shales. In the following years (Cobme,
17
1932) economic quantities of alluvial gold were found in south of Lubare ridge and a deposit was opened
by a private company on the Buhweju plateau (Wayland, 1934). Various compilation works on the mineral
resources of Uganda showed that Buhweju was still the main gold producer came entirely from wide
spread alluvial mining where the source rock is still unknown. Alluvial of Karagwe-Ankolean rocks occur
in streams, stream courses and swamps with heavily vegetated surroundings. Different observations have
also indicated that the gold was mostly found on the lower parts of the gravel and couldnt be detected in
the bed of the stream. However, at some areas like Kanyambogo and Kitaka gold has been mined from
the veins cutting the basement Igara schists. According to Wayland (1934) reef gold has also been found
in Muti area forming stockwork of quartz strigers transecting a quartzite bore of pyrite and fine gold. He
has also showed the gold occurrences in the sandy transition of the quartzite and pelites. Three sulphide
veins have also been identified (Reece, 1961) at Kitaka, Kampono and Kanyambogo lead-zinc mines
emplaced in shcists and gneiss. In these mines coarse crystalline gold also occur in vughs lined by quartz
crystals. In Kitaka mine galena, chalcopyrite and gold have been carried by undulating quartz veins
emplaced in shears within metadolerite. According to the explanation by Reece (1961) gold mineralization
is in general epithermal as is shown by its often coarsely crystalline occurrence in vughs. As inferred from
the previous study by Pohl (1994) mineralization has been started ca 1250 Ma related to post orogenic
rifting. Mineralization at ca 950 Ma however considered as dominant which has been related to post
orogenic metamorphism.
18
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
19
(Figure 3-2) and used for further enhancements and interpretations. Most of the subsequent image
enhancement and analysis were made on the eastern part of Buhweju due to its complexity covered by
Precambrian Buhweju and Igara groups as compared to the western part, covered by recent volcanic and
sediments.
Figure 3-1: NDVI Images of the study area A) ASTER image B) Landsat ETM+
20
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
A positive correlation has observed between the nine ASTER VNIR-SWIR bands (Table 3-2). The first
three bands especially band 1 showed very low correlation with the other bands except band 2 (0.81).
Therefore an appropriate triplet bands in RGB 741, 731 and 721 (Figure 3-2) are selected which gave
better color composite for visual interpretation. Generally, ASTER741, ASTER721, and ETM751 were
selected for further interpretations of lithology and structures since they showed better reparability in
geology.
Band 1 Band 2 Band 3 Band 4 Band 5 Band 6 Band 7 Band 8 Band 9
Band 1 1.00
Band 2 0.81 1.00
Band 3 0.17 0.53 1.00
Band 4 0.29 0.75 0.87 1.00
Band 5 0.33 0.78 0.79 0.98 1.00
Band 6 0.37 0.81 0.78 0.98 1.00 1.00
Band 7 0.33 0.78 0.78 0.98 1.00 1.00 1.00
Band 8 0.34 0.79 0.76 0.97 0.99 0.99 1.00 1.00
Band 9 0.24 0.73 0.78 0.97 0.98 0.98 0.99 0.99 1.00
Table 3-2: Correlation coefficient of ASTER VNIR-SWIR data of the area
21
3) were used for better extraction of lithological information (Figure 3-4) since they use bands with highest
variance and least redundancy (Qaid & Basavarajappa, 2008).
Figure 3-2: Subscence of Landsat ETM and ASTER color composite images in (RGB) A)ASTER731 B)ASTER721
C)ASTER741 D)ETM751
Figure 3-3:Subscence of OIF color composite in RGB order A) ASTER TIR13 VNIR2 VNIR3 B) ASTER VNIR1
TIR13 VNIR3 C) ETM742 D) ETM431
22
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
23
Eigen Band1 Band2 Band3 Band4 Band5 Band6 Band7 Band8 Band9 Variance
vector (%)
PC1 0.21 0.34 -0.03 0.34 0.39 0.38 0.38 0.39 0.36 79.75
PC2 0.44 0.30 0.74 0.24 -0.80 -0.10 -0.12 -0.19 -0.22 8.77
PC3 -0.42 -0.63 0.54 0.24 0.13 0.10 0.14 0.11 0.14 6.77
PC4 -0.10 -0.1 -0.35 0.76 0.10 0.11 -0.10 -0.26 -0.45 1.9
PC5 -0.60 0.52 0.10 0.26 -0.32 -0.20 -0.15 -0.05 0.37 0.92
PC6 -0.46 0.35 0.15 -0.31 0.50 0.13 0.10 0.03 -0.55 0.65
PC7 0.15 -0.03 -0.03 0.00 0.70 -0.21 -0.40 -0.05 0.40 0.52
PC8 0.10 -0.03 -0.04 0.18 0.21 -0.70 -0.21 0.63 -0.16 0.37
PC9 -0.02 -0.02 0.02 -0.03 -0.03 0.53 -0.80 0.32 0.00 0.36
Table 3-4: Eigen vector loadings of principal components for ASTER VNIR-SWIR data and their variance
Eigen vector Band1 Band2 Band3 Band4 Band5 Band6 Variance (%)
PC1 0.13 0.13 0.15 0.12 0.13 0.96 98.88
PC2 0.67 0.36 0.48 0.26 0.20 -0.28 0.68
PC3 0.71 -0.27 -0.43 -0.29 -0.38 0.10 0.32
PC4 0.12 -0.21 -0.55 0.05 0.61 -0.04 0.09
PC5 0.06 -0.85 0.50 0.00 0.15 0.01 0.03
PC6 0.06 0.10 -0.01 -0.76 0.64 -0.01 0.01
Table 3-5: Eigen values of the principal component of ASTER SWIR data of the area.
Figure 3-4: Subscens with interpreted lithologies A) ASTER VNIR-SWIR PC1 B) ASTER VNIR-SWIR
PC2 Sc=schist Gr=granite Gn=gneiss Pl=pelite Qzt=quartzite Und=Undifferentiated unit
24
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
Table 3-6: Eigen values of principal components of Landsat ETM data and their variance
Figure 3-5:. Subscens with interpreted Lithologies A) PC1 and B) PC3 of Landsat ETM+ Gr=Granite
Gn=Gneiss Sc=Schist Pl=Pelite Qzt=Quartzite Und=Undifferentiated granite, schist and gneiss. Location
see inset map in (Fig. 3-4)
25
3.5. Summary of multispectral processing and interpretation
The complex topographic features, less variability in reflectance of the various lithologies and vegetation
cover have greatly influenced the amount of geological information that could be obtained from the
enhanced composite images. Correlation coefficient, OIF and PCA techniques were applied to both
ASTER and Landsat ETM+ data. Transformations of color space to IHS and decorrelation stretch were
also performed on various composite images but interpretation became difficult due exaggerated colors.
Consequently, the different PC combinations and band ratios couldnt be used to enhance the various
rock types composed of mainly of iron, carbonates or hydroxyl minerals. However, the univariate and
PCA analysis have resulted better enhanced images for photo geological interpretations. Combinations of
color composite image including ASTER721, ASTER741 and ETM751 were found considerably useful
for lithological and structural interpretations of the area. In addition the various single PC images
including PC1 (albedo) images were used to interpret surface lineaments of the area. The multispectral
interpretation of the area has discriminated some lithologies which were mapped as same unit by (Reece,
1961). For example, the previously mapped undifferentiated granite, gneiss and schist have separated in to
two different units based on the absence and presence of drainages. Accordingly, a generalized discussion
on geology and an interpreted lithological map (Figure 3-6) of the area were made from the analysis
presented above with a comparison of previous study and field work (Table 3-7).
26
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
27
4. SURFACE LINEAMENT INTERPRETATION
4.1. Introduction
Different definitions have been given since early times for the commonly used term in geology called
Lineament. However, most researchers have agreed that lineaments are mappable linear or curvilinear
surface features which distinctly differ from adjacent patterns and presumably represent topographic
expressions and subsurface phenomenon (Gupta, 1991; Masoud & Koike, 2006; OLeary et al., 1976;
Williams, 1983). Lineament extraction from multispectral imageries has been used extensively basically to
determine the structural grain of a region and for various other purposes like mineral exploration.
Multispectral satellite images are considered to be better tool in discriminating lineaments since they are
obtained from varying wavelength intervals of the electromagnetic spectrum (Casas et al., 2000).
According to (Abdullah et al., 2010; E. A. Owusu et al., 2006; Sarapirome et al., 2002; Singh & Dowerah,
2010; Soulakellis et al., 2006). Manual interpretation and automatic extraction are the two commonly used
methods to extract lineaments from multispectral images. In this study visual extraction method was
applied on ASTER, Landsat ETM and SRTM DEM images to discriminate lineaments of the area.
Elevation data provides the topographic expressions of an area and have been applied for structural
interpretation by various investigators in the past (Henderson et al., 1996; A. E. Owusu et al., 2006;
Soulakellis et al., 2006). Elevation data can be represented in different forms among which the grid forms
(DEMs) are preferably useful for interpreting linear geologic structures which have topographic
expressions due to offset of the surface. The ability of producing different shaded relief images because of
the freedom to select illumination from any angle (Henderson et al., 1996) makes DEM selectively useful
over Arial photographs (Abdullah et al., 2010; E. A. Owusu et al., 2006; Sarapirome et al., 2002; Singh &
Dowerah, 2010; Soulakellis et al., 2006). According to the authors, solar elevation and azimuth are
important elements to examine topographically related and dependent features. DEMs could also be fused
with other multispectral images to increase interpretability and obtain more information than can be
derived from each of the single image alone (Soulakellis et al., 2006). The various image enhancement
techniques including directional, edge and soble filters can be applied to DEM to enhance different linear
structures (Henderson et al., 1996). Therefore, extraction of surface lineament, assess their general pattern
and density to decipher structural information of the area, is the main aim of the chapter.
28
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
Figure 4-1: Subscenes of A) ASTER 721 in RGB order B) Landsat ETM5 overlaid by with extracted lineaments
Table 4-1: Soble filter in four main directions applied in this study
29
4.1.2. SRTM DEM interpretation
The SRTM DEM of the Buhweju area having a 33m resolution provided good topographic and structural
information for use in lineament analysis. The extraction of lineaments from SRTM DEM in the Buhweju
area involved three steps: (1) shaded relief images were produced, (2) shaded relief images were fused with
multispectral data and (3) linear vector features were extracted from the processed images. Shaded relief
images were produced (Figure 4-3) from eight different Azimuth directions namely N, NE, E, SE, S, SW,
W and NW. These relief images were helpful for enhancing the size, height, and slope variations of a
morphology. Different sun angles like 00, 150, 300 and 450 were tried to get more information among
which 300 was finally selected for its better interpretability. Figure 4-3 shows the images produced by
combining the different shaded relief images facing to the same direction. The shaded relief images with
illumination directions N, NE, E were stacked and stretched to enhance E-W, N-S and NW-SE
lineaments (Figure 4-3a). The other shaded relief image was produced by combining N, NW and W
direction relief images to enhance E-W, N-S and NE-SW lineaments (Figure 4-3b.) These shaded relief
composite images in general led to better interpretation of lineaments since they superimpose the
information contained in various single shaded relief images.
Fused images were generated by multiplying the landsat ETM band5 with the eight shaded relief images
produced before. Band5 was selected because of its better information on topography and boundary
delineation as compared to other bands. From the multispectral datasets describe in (section 3.3)
ETM741 color composite offered better delineation of lithologic units in comparison to lineament
extraction. Good lineament interpretation however was possible when transformed to IHS and Intensity
enhanced by the color shaded image (Figure 4-4).
The final interpreted lineament map (Figure 4-5a) was then produced with a total of about 248 lineaments
extracted from SRTM DEM. The azmuthal distribution (Figure 4-5b) indicated that most of the
lineaments are trending to NNW-SSE and NNE-SSW direction directions and distributed though out the
Buhweju and Igara complexes.
Figure 4-3: Color shaded relief images. A) Composite SRTM image with illumination directions of 45, 90, 360
B) Composite image of SRTM with an illumination directions of 270,315 and 360 degrees.
30
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
Figure 4-4 :SRTM DEM Subscens of the area A) SRTM DEM with illumination angle at 450 B) IHS image of
Landsat ETM741 in RGB, Intensity modulated by SRTM DEM. White arrows indicate enhanced linear features
trending in different directions.
31
4.2. Summary
From the above sections it has observed that almost the same number of lineaments was extracted from
both interpretations. However, there are lineaments which were identified in multispectral data but not in
SRTM DEM and vice versa. Therefore, the final integrated lineament map of the area (Figure 4-7) was
produced by synthesizing these preliminary interpretation maps. The following interpretations were
generally made concerning the surface lineament of the area:
1. As clearly seen from the SRTM DEM, the central part of the area (Buhweju complex) has a
remarkable elevation difference as compared to adjacent areas. This area is entirely covered with
Precambrian Buhweju group rocks consisting of pelites and psammites intimately mixed with
quartzite. The lineament interpretation has showed that the area is topographically complex,
highly dissected and hence high lineament density (Figure 4-8) has observed.
2. The Lubare quartzite has showed a prominent emplacement bounding the southern part of
Buhweju. This unit has also dissected by several strike slip faults of various (N-S, NNW-SSE and
NE-SW directions.) throughout its length from west to east with alternating dextral and sinistral
movements. The Lubare quartzite at the northern part however is dissected by lineaments with
NW-SE and E-W orientations.
3. The western part of the study area partly constitutes eastern portion of the western rift valley and
hence different NE-SW trending stepping normal faults are clearly observed crossing the recent
rift volcanic and sediments.
4. The density lineament map was revealed that the Buhweju group rocks especially the Lubare
quartzite was highly affected by high lineament concentration.
5. Two prominent lineament orientations are interpreted from the rose diagram. These are
lineaments trending in about N10-50E and N0-40W both occurring mainly on the Buhweju
complex and the former mainly attributed to the rift related faults.
32
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
33
5. PROCESSING AND INTERPRETATION OF GAMMA
RAY DATA
5.1. Introduction
Nowadays, gamma ray spectrometric data are becoming popular method for surface geological mapping
and detection of alteration related to mineralization. Basically, the method measures the abundance of
most common radioactive elements, Potassium, Thorium and Uranium for the top 0.30 m of the Earth's
surface. Mapping the surface geology through interpretation of radioelement distribution relies on the
assumption that rocks are composed of rock forming minerals containing specific amount of radioactive
elements. In addition, the capability of surficial materials to retain significant and detectable compositional
difference between lithologic units determines the efficiency of gamma ray for surface geology mapping
(Dickson & Scott, 1997; Wilford, 2002). Various previous works have already been published on
utilization of gamma ray data for regolith and surface geology mapping (Dickson & Scott, 1997; P. J.
Gunn et al., 1997; Jayawardhana & Sheard, 2000; Johnson et al., 1979; Wilford, 2002; Wilford et al., 1997;
ZHANG et al., 1998) and for identification of alterations related to mineralization (Jayawardhana &
Sheard, 2000; Quadros et al., 2003; Ramadan et al., 2009; Shives et al., 2000).
There are several factors which affect the merit of gamma ray data for geological mapping including;
vegetation cover, detectable contrast in radioelement between lithologic units, weathering and soil
moisture. Despite of these controlling factors however,, gamma ray data has showed advantages over
other remote sensing data set for mapping the surface lithology in vegetated and weathered areas (Perrotta
et al., 2008). Gamma ray response from arid and unweathered terrain is directly related to mineralogy and
geochemistry of bed rock. Hence, this good correlation usually makes interpretation less difficult. In
tropical areas like Uganda, where vegetation, weathering and geomorphic processes are significant
however, radiometric responses are usually a difficult combination of responses from various surface
covers including fresh rock, weathered rock and transported materials (Dickson & Scott, 1997).
Weathering greatly affects the distribution and concentration of radioelement and hence understanding
how radioelements behave during weathering would be crucial to make an interpretation (Wilford, 2002).
The geochemical characteristics of weathered rocks are different as compared to the rocks from which
they were derived. This is because the intensive weathering over a long time have leached and transported
the soluble elements and have left behind the insoluble components (Butt et al., 2000). According to
Dickson & Scott (1997), Gunn et al. (1997), Wilford, (2002) and Wilford et al. (1997) therefore, a general
assumption could be drawn that in most rock types K concentration decreases with increasing degree of
weathering (Figure 5-1a) which also works true for U and Th but in felsic rocks only. Weathering of
intermediate, mafic and ultramafic rocks produce soils with relatively increased U and Th concentrations.
This is because K leaches very easily due to its highly soluble nature whereas U and Th are mostly
associated with resistant minerals and are scavenged by the weathering profile during weathering. The
study by Wilford et al. (1997) has shown that, geomorphic activities can also play a significant role in
radiometric signature. This implies that an actively eroding terrain reflects the bed rock response on
radiometric data whereas stable terrains reflect regolith and weathering responses (Figure 5-1b).
34
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
Figure 5-1: A) Element weathering and gamma ray response after (Wilford et al., 1997) B) Relationship
between gamma ray response and denudation balance in landscapes(Wilford, 2002)
Many previous geological remote sensing studies have highly recommended and demonstrated that
gamma ray data should be integrated with other remote sensing data for better knowledge of
interpretation (Dickson & Scott, 1997; Perrotta et al., 2008; Wilford et al., 1997). The geomorphic
activities and the corresponding gamma ray response will be well interpreted when integrated with
topographic data (DEMs). In addition, integration of gamma ray data with satellite images of better spatial
resolution have resulted better geological interpretation because the combined image enables the
geochemical properties contained in gamma ray to be interpreted in terms of actual surface features
(Dickson & Scott, 1997; Schetselaar et al., 2000; Schetselaar et al., 2008; Wilford et al., 1997).
To observe the range of concentration of each radioelement over the area and to calculate the correlation
between the radioelements the following procedures were followed. First, a separate geo database was
created for each element in Oasis Montaji which consists of latitude-X, longitude-Y and concentration-Z.
Second these data were exported to xyz file for further statistical analysis. The approximate range of
concentration of the radioelements in the area is 0.05-2.89 %, 0.1-40.5 ppm and 0.5-10 ppm for K, Th and
U respectively. The correlation factor between the radioelements are also calculated and shown in the
(Table 5-1) indicating that generally high correlation exists between Th and U.
35
Table 5-1: Correlation between radioelements
The relation of each of the radiometric elements with topography is rather complex but generally showed
that elevation is directly proportional to Th and U concentration and inversely related to K concentration.
Statistical analysis was also carried out to depict the concentration of each radioelement per various
lithological units of the area. The mean values of each radioelement concentrations per lithological units
(Figure 5-2) suggested that it is possible to discriminate geological units based on their radiometric
signature mainly of K and Th.
Figure 5-2: Box plot showing the distribution of radioelement per litholohic units A) K-plot B) Th-plot C) U-plot D)
Key to the numbers of mapped lithologic units in A, B and C
36
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
Where a, b and c are constants (a=1; b=2) used to weight the contrast between the landsat ETM band 5
and radioelement channels to increase or decrease the influence of one image with respect to the other. In
addition, C should always be the sum of a and b to obtain values which are in the same range as the
impute images.
37
Likewise, the radiometric data were also integrated with the STRM DEM data to add topographic
information. After co registering to the same grid cell size, the SRTM DEM data was combined with the
radiometric data basically by two different ways. The first approach involved the transformation of RGB
radiometric composite image (ternary) into Intensity, Saturation and Hue (IHS) which is more intuitive for
human color vision. Once RGB is transformed to IHS, there is a possibility to substitute the different
channels by another data as required. The intensity image was substituted by the SRTM DEM and an
inverse transformation of IHS image to RGB was carried as seen in the (figure 5-5b). The second
approach involves pan-sharpening; the low (50m) resolutions radiometric data is fused with a high
resolution (33m) STRM DEM (Figure 5-6c). In addition, the composite image (ternary) was enhanced by
SRTM DEM with hill shading in different directions (Figure 5-6d) using Oasis montaj. This has resulted
in a better enhanced image in comparison to the original ternary image. The IHS image obtained from the
RGB ternary image was not good as such and not used for further interpretation and analysis.
Figure 5-3: Images of individual radioelement grids A) K grid B) Th grid and C) Ugrid
Figure 5-4: Images radioelement ratio grids A) Th/K map B) U/K map and C) U/Th map
38
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
Figure 5-5: Various IHS composite images of gamma ray data, K, eTh, eU in RGB order A) The
Intensity data is substituted by Landsat ETM5. In B) the Intensity is substituted by SRTM DEM in
C) by PC1 image of ASTER VNIR-SWIR and in D) by Landsat ETM7
39
Figure 5-6: Different enhancements made on Ternary image A) Original ternary image KThU in RGB
B) Ternary image modulated by shaded relief of total count C) Ternary image pan-sharpened by SRTM
DEM D) Ternary image modulated by SRTM DEM with illumination from the 1st quadrant ( 450 ).
40
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
5.3.2. Interpretation
Qualitative photo geological interpretations were applied here similar to interpretations made in section 3
on multispectral images.. Lithological units were discriminated based on their Tonal/ color difference with
blue colors representing low and red representing high concentration of a particular radioelement. The
composite images including ternary one were used to display the distribution pattern of the three
radioelements (K, eTh, eU) in red, green and blue respectively.
Close inspection was made on the enhanced images produced in section 5.3.1 in comparison with the
previous geological map produced by (Reece, 1961). There was generally good agreement between the
radiometric signature and the mapped lithologic units except mismatching of lithological boundaries.
There are distinct lithologies which can easily be discerned from K and Th channel. These include: The
Pleistocene tuffs and agglomerates located in the NW corner; the Precambrian units including pelitic
schist in the SW; pelites with minor quartzite located in the central part and gneissic terrains in the SE part
of the study area (Figure 5-7a). In addition, interesting radiometric distributions were obtained from the
ratio images mainly Th/k and U/K from which Th and U enriched lithologies were clearly identified with
respect to K concentration. The Precambrian lithologies covering the western part of the area (Figure 5-
7b) including the undifferentiated granite, gneiss and schist, Quartzite and gneiss in the SE part have
showed high concentration of Th and U while the pelitic schist, pelites (with minor quartzite) including
some of the areas previously mapped (Reece, 1961) as undifferentiated unit showed low Th/K ratio value.
As shown in the (Figure 5-7b) the undifferentiated granite, gneiss and schist mapped by (Reece, 1961) has
showed high and low Th concentrations in the east and west of the Lubare quartzite respectively.
Interpretations were also made on the various enhanced composite images for example ternary (Figure 5-
8a) and K-composite images (Figure 5-8b). Ternary image discriminate best between the different
lithologic units of the area. The lubare quartzite, gneiss, pelitic schist and pelites (with minor quartzite) are
clearly identified from the ternary image. The gneissic terrain at the southern part showed higher eTh and
eU concentrations displaying cyan color on the image. The lubare and Ibanda quartzite are also
discriminated from other lithologies by their elevated eTh concentration appearing green to bluish green
on image. Pelites are clearly identified having white color which indicates higher concentration of K, eTh,
and eU. This unit became reddish in color going from east to west which could be due to effect of
vegetation that attenuates the penetration power of radiation from radioelements and hence decreases the
concentration of eTh and eU. The boundary of this unit was clearly outlined also in K composite image
as shown in Figure 5-8b. The quartzite band dominated by psammitic units showed increased K as
compared to other radioelement which could be related to the composition of psammitic units. The above
interpretations made clear that the undifferentiated granite, gneiss and schist has apparently classified in to
units having elevated K and eTh as shown in (Figure 5-8a and Figure 5-8b).
The fused and sharpened images (Figure 5-9a and Figure 5-9b) have confirmed most of the above
interpretations extracted from the radiometric data. The intensity enhanced IHS image (Figure 5-9b) has
provided clearer and sharp lithology contact and drainages with their radiometric information. The image
also showed the effect of vegetation and topography on radioelement distribution of the area which
suppressed the concentration of mainly eTh and eU. Lithological and other structural features are
apparent and easily identified on the ternary image pan-sharpened by SRTM DEM. This image also
revealed the distribution of radioelement with topography of Buhweju.
41
Figure 5-7: Previously mapped lithology overlaid on A) Th map and B) Th/K map Pl=Pelite with minor
quartzite Gn= Gneiss Sc=Schist SCC=Sand, silt and clay Lqzt=Lubare quartzite Tf= Tuff and
agglomerates Udf= Undifferentiated granite, gneiss and schist. Black lines represent lithologic boundaries
by (Reece, 1961)
Figure 5-8: Lithological interpretations on gamma ray data A) Ternary Image B) K-composite image. White lines
indicate previously mapped lithology boundaries by (Reece, 1961) and Black line represent interpreted lithology
boundary. Gn=Gniess Sc=Schist Pl=Pelite with minor quartzite Lqzt=Lubare quartzite Scc= Silt,clay sand
Tf=Tuff and agglomerates
42
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
Figure 5-9:Interpreted lithologic units from radiometric data A) Ternary image pan sharpened by SRTM
DEM B) IHS Composite image, Intensity enhanced by Landsat ETM5. Black and white lines represent
interpreted lithology boundaries. Pl=Pelite Gn=Gniess Sc=Schist Lqzt=Lubare qaurtzite Qzt=Quartzite
with minor pelite dominantly psammitic Scc= Sand,cilt,clay Tf=Tuff and agglomerates Gr=Granite
Cgl=Conglomerates and grits
Figure 5-10: Interpreted lithological map. Numbers indicate the newly identified units
43
5.4. Summary of Radiometric data interpretation
Several image enhancement techniques were applied including single radioelement maps, ratio maps,
image fusions and pan-sharpening of radioelement with multispectral and SRTM DEM data sets. Most of
these enhancements have increased the interpretability of the radiometric data especially of the pan-
sharpened and IHS transformed images.
Radiometric data generally found better in delineating and discriminating the different lithological units of
the area. Most of the previously mapped lithological units (Reece, 1961) are in good agreement with the
prevailing radioelement maps. The radiometric data interpretation revealed new lithologies which were not
identified by Reece, (1961). These are schist and gneiss, identified from the previously mapped
undifferentiated unit located in the northern part of the area, granite identified from the gneissic terrain in
the SE of the area. The gneiss and pelitic schist in the SW have also showed variation in concentration of
K which might be due to variation in degree of weathering throughout the unit. However, the names
given for the newly identified units are entirely depending on their previous name by Reece, (1961).
Finally, lithologies identified from analysis and interpretation of radiometric data is presented in the map
(Figure 5-10) and their radiometric correlation is shown in (Appendix-2). The radiometric correlation for
the newly identified units has only presented here in the table below.
New
Interpreted Description and Radiometric signature
units
Schist -1 Compositionally rich in muscovite and biotite mica. Quartz is also common. Previously
mapped as undifferentiated granite, gneiss and schist. Radiometric signature showed
elevated K/Th value and very low Th/K ratio value. It appears dominantly red on
enhanced ternary maps.
Schist -4 Compositionally the same with schist (1) and also mapped as schist by (Reece, 1961). An
increased K concentration is observed on most radiometric images. It appears green in K
composite images implying an increased K/Th ratio value.
Gneiss -2 and Mainly composed of quartz and feldspar with occasional muscovite and mapped as
5 undifferentiated granite, gneiss and schist by (Reece, 1961). Gneiss (2) has showed the
same signature with other gneissic terrains in the SW corner (Figure 5-10) which
appeared cyan due to enrichments in eTh and eU. Whereas in gneiss (5) depletion in Th
and U was observed probably due to less weathering condition towards south.
Granite -3 and Compositionally composed of quartz and feldspar with occasional muscovite. Granite
6 (3) appeared dominantly white in ternary image due to higher concentration of
radioelement. Granite (6) has mapped before as schist by (Reece,1961)
44
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
45
pole. RTP positions the anomaly directly above the source and hence makes interpretation easier. For this
reason, other layers of information are derived from the RTP of total magnetic intensity (TMI) grid. The
enhanced RTP image of the area (Figure 6-1a) showed the magnetic field amplitude of about 110.4nT
indicating the variation of the magnetic intensity due to either lithological or topographical changes.
Analytical signal was calculated for the area (Figure 6-1b) which positions the anomaly at the centre of the
causative body by combining both vertical and horizontal derivatives. Analytical signal is considered better
enhancement technique since dipolar effects are absent and even for small bodies the peaks merge
resulting an anomaly cantered above the causative body (Alsaud, 2008). It is also observed that the vertical
derivative applied to the analytical signal has sharpened up and positioned the anomaly more exactly than
the original AS image. The tilt derivative (Figure 6-1c) applied to the RTP data has also provided better
information by positioning the anomalies above the causative body like RTP and analytical signal.
Horizontal gradient image was calculated from the RTP and upward continuations were performed at
various depths to depict deep seated anomalies and magnetic contacts. Upward continued to 1km was
selected for further interpretation as shown in (Figure 6-1d). The tilt derivative accentuates short wave
length anomalies and was found to be effective in allowing anomalies to be traced along their strike
(Alsaud, 2008). Both first and second vertical derivative images (Figure 6-2) were found better for
delineation of shallow surface magnetic structures. Various angles of illuminations were applied on AS,
RTP and tilt derivatives to produce shaded relief images which emphasized different sets of structures. In
addition, several ratio and composite images (e.g. RTP/AS, RTP/VD1, and VD1/AS) were used to
constrain interpretations.
6.2.2. Interpretation
46
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
Precambrian basement, pelite and psamitic rocks were dominantly observed together with their dominant
orientation (Figure 6-1d). Finally, high magnetic zones were extracted from interpretations made on the
above layers as presented in the figure 6.3b.
Figure 6-1: Enhanced Aeromagnetic images of the area A) Reduced to the pole (RTP) of TMI B) Analytical signal
B) C) Tilt Derivative D) Horizontal gradient upward continued to 1km Black lines: Lithology boundaries Letters
indicate identified magnetic faults. For detail see text.
Directions of anomalies mainly depend on orientation of magnetic sources which in turn could be
tectonically controlled. Analytical signal, upward continued horizontal gradients and tilt derivative were
found particularly valuable in amplifying orientations of magnetic anomalies. These anomaly orientations
were then extracted and analysed separately (Figure 6-3) to decipher the tectonic grain of the area.
Accordingly statistical analysis was carried out to determine the azimuthal distribution and result showed
that the major magnetic anomalies are mostly trending in two prominent directions. These are N0-40 W
and N0-50E as shown in the figure 6-3b.
47
Figure 6-2: Vertical derivative images calculated from RTP A) First vertical derivative B) Second vertical derivative
Figure 6-3: A) Regional litho-magnetic domains interpreted from RTP. B) High magnetic anomaly
zones. Rose diagram shows the azimuthal distribution of orientation of magnetic
48
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
Figure 6-4: Interpreted structural map of the area. Rose diagram shows the azimuthal distribution of the
magnetic lineaments.
49
Figure 6-5: Magnetic lineament density map
50
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
51
7.2.2. Pelitic Schist
Previously this unit has been mapped distinctly in the south and with intimate mixture of granite and
gneiss in the north and east. Field investigation revealed that the unit is compositionally rich in mica with
gradation in quartz. Radiometric data was found valuable in discriminating the unit from other lithological
units due to its elevated K concentration (Figure 7-1b) and very low Th/K ratio. On the basis of similarity
of this radiometric signature and in contrast to the previous geological map this unit was also identified in
the northern part which has been previously mapped as undifferentiated granite, gneiss and schist (Figure
7-1d). The pelitic schist in the south has showed sharp enrichment of K towards its north which might be
due to variation in weathering and hence this study has mapped the unit separately. In the south the pelitic
schist has showed distinctive magnetic signature revealing short wavelength curvilinear anomalies which
defined folded rock foliation. The newly identified unit in the north however was characterized by the
emplacement of both deep and shallow seated high magnetic rounded bodies. Detail explanation on
radiometric signature of units is presented in (section 5.3.2).
Figure 7-1: Subscens showing interpreted lithological units A) Old geological map by Reece, 1961 B) Ternary image
C) Improved geological map by this study D) Interpreted lithological units from gamma ray data interpretation
7.2.3. Gneiss
The unit has been previously mapped in the SE and SW corners of Buhweju (Figure 7-2a). The previous
work by Reece, (1961) has indicated the intimate association of gneiss with granite and minor pegmatite. It
was difficult to characterize this unit in the field due to prolong intense weathering of the area but
compositionally mainly made of quartz and feldspar with minor muscovite. However, radiometric data
was found valuable in discriminating the gneissic terrains of the area through revealing higher
concentration of eTh and eU which appeared cyan color on ternary image (Figure 7-1b and 7-2b). In
contrast to the previous geological map more gneissic units were identified in the eastern and northern
part of the area on the basis of similarity of radiometric signature. For example the area under the Ibanda
quartzite (Figure 7-1b) has showed distinctive radiometric similarity with the already mapped gneissic
terrains found in the SE and SW corners of the area. In addition, this study has identified the variation of
K concentration throughout gneiss mapped at the SW corner (Table 5-2). This might be due to the
52
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
variation in degree of weathering within this gneissic terrain. The aeromagnetic data has indicated that this
unit has underlain by both shallow and deep seated high magnetic signatures (Figure 7-2c).
7.2.4. Granite
Reece, (1961) has mapped small units of granite in the south and east of Buhweju (Figure 7-2a). He has
also noted that the unit has found intimately mixed with gneiss, pegmatite and aplites. However field data
has showed relatively wider area covered by granite than mapped before especially in the southern part of
the area. The intimate mixture of the unit with gneiss has also proved by very close similarity in
radiometric signature. However, close inspection on radiometric composite ternary image has depicted the
presence of granitic signature (Figure 7-2b). Aeromagnetic data didnt exhibit considerable variation
beneath pelitic schist, gneiss and granite units exposed in the southern part of Buhweju (Figure 7-2c).
Therefore, a wider area was identified as granite in the southern part which has been previously mapped as
gneiss (Figure 7-2d).
Figure 7-2: Subscenes showing interpreted lithological units A) Old geological map by Reece, 1961 B) Ternary image C)
Analytical signal image D) improved geological map by this study
7.2.5. Quartzite
Two different quartzite types have been mapped by Reece, (1961): the Lubare quartzite and the Ibanda
quartzite (Figure 7-1a). The relationship between Lubare quartzite, gneiss and schist has not yet been
established. Previous study and field data indicated that both units were compositionally dominated by
pure quartz with minor feldspars and muscovite. Two conspicuous characters were observed at both
quartzite units: elevated topography and bounded by conglomerates and grits. This uplifting of
stratigraphicaly lower unit to its present position might be tectonically controlled. Both units have
indicated similar radiometric signature with elevated eTh concentration and hence appear greenish on
ternary image (Figure 7-1b and 7-2b). Generally low aeromagnetic signature was depicted underneath
quartzite unit.
53
7.2.6. Pelite and psammites
These units have covered large area of the Buhweju group rocks covered by dense protected forests and
intimately mixed with quartzite. The boundary of these units was identified in most of the data sets used in
this study. Generally elevated K, eTh and eU concentrations have been observed in pelite (Figure 7-2b).
The presence of dense vegetation cover to the west would be the plausible reason for the gradual decrease
in concentration of radioelement towards west. The quartzite bands (dominated by psammitic units) on
the other hand showed relatively less concentration of radioelement and seemed to be affected by the
topographic effect (Figure 7-2b). Two distinguishable units were depicted from visual interpretations
vegetation texture and pattern applied on color composite, PCA and single bands of multispectral data
(section 3.5). Broad, smooth and long wavelength magnetic anomaly was exhibited underneath these
lithological units which proved the presence of deep seated basement (Figure 7-2c). The eastern part of
these units has prominently intruded by a sub surface dyke with NNE trend. In addition, the boundary of
pelite (with minor quartzite) has marked by elongated high magnetic signature. The quartzite (dominated
by psammites) on the other hand has specifically underlain by shallow seated rounded magnetic anomalies.
Figure 7-3: Field photos of different rocks A) Weathered granite B) Lubare quartzite C)
Gneiss D) Folded tuff
54
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
55
7.3. Compiled Lineament Interpretation
Detail surface and subsurface lineament interpretations and analysis have been carried out on multispectral
data, SRTM DEM data and aeromagnetic data (chapter 4 and chapter 6). The surface lineament density
analysis (Figure 4-7) has depicted that the Buhweju group rocks (pelites, Psammites and Lubare quartzite)
were affected by high lineament density along a narrow zone. It has also observed that this high density
lineament continued further to the north following a narrow zone trending NNE and underlain by
subsurface fault with high magnetic anomaly. The density analysis applied on subsurface magnetic
lineaments (Figure 6-5) revealed high lineament concentration affecting the northern undifferentiated
schist and amphibole including the newly identified pelitic schist. The Buhweju group rocks were also
influenced by the subsurface magnetic lineament however the correlation between surface and subsurface
lineament was generally considered weak. The orientation analysis has proved the existence of three
tectonic patterns namely: NNW-SSE, ESE-WNW and NNE-SSW. These tectonic patterns were found
consistent with the regional tectonic grain. The first two patterns are consistent with the regional Aswa
shear zone (Shackleton, 1976) and the refolding event of both Igara and Buhweju Group rocks (Reece,
1961) respectively. Lineaments and faults trending NNE-SSW were considered to be coherent with the
general trend of the western rift valley (Riad & El Etr, 1985). The orientation analysis applied on the
quartz veins has resulted two prominent directions: N50-80E and N20-50W. The Lubare quartzite in the
southern part is affected by sinistral movements as depicted from the multispectral, DEM SRTM and
horizontal gradient images. The interpreted lineament map of Buhweju is shown in figure 7-5.
7.4. Mineralization
The regional and local aspect of mineralization in Buhweju area was discussed in detail (chapter 2). Post
Kibaran orogeny events have been considered the prime cause for most of Kibaran mineralization.
Generally two episodes of mineralization have been recognized: mineralization related to post orogenic
rifting and mineralization related to post orogenic metamorphism (Pohl, 1994). The majority of gold
occurrence in Buhweju has been from alluvial deposits but reef gold associated with sulphide veins has
also been found (Reece, 1961; Wayland, 1934). The aim of this section is therefore to assess the control on
gold mineralization in Buhweju area by synthesizing field data, mineral occurrence data, and interpreted
lithological and structural data. The field surveyed gold mine areas and the provided gold occurrences data
(Appendix 1) are used primarily for this purpose. Field data and lithological interpretations (Figure 7-6a)
have revealed that Lubare quartzite and mudstone are the dominant country rocks hosting gold
occurrences of the Buhweju plateau. Gold occurrences found below the plateau were hosted by different
bedrocks including schist, gneiss, granites and amphibolites. In the surveyed mine areas, for example
Kitaka mine suphides which occur in quartz veins have provided gold and found disseminated throughout
the host rock (Figure 7-6b). As shown in (Figure 7-6c) most of the alluvial gold mines are located in low
land stream courses covered by vegetation.
Spatial association has observed between surface lineament density and gold mine areas where the high
lineament density and NNE tending narrow zone was observed hosting most of the gold mine areas. As
shown in (Figure 7-7a) most of the gold mine occurrences are densely clustered in two areas. These
clustered gold occurrence areas have showed difference in lithologies and subsurface magnetic signature.
The gold occurrences at Kitaka mine are hosted by schist (Figure 7-7b) and underlain by low magnetic
material (Figure 7-7c). On the other hand the gold occurrences at Katonga swamp are hosted by Lubare
quartzite and mudstone (Figure 7-7b) and underlain by relatively high magnetic material (Figure 7-7c).
There is also an association observed between the gold occurrences and quartz veins as shown in the
figure 7-7d.
Currently exploration works are being in progress at different gold fields found in Tanzania, Uganda and
Democratic Republic Congo (DRC). Two major strucutural corridors (gold trends) have been already
identified by Magnus and Banro exploration companies (Corporation, 2006; Magnus, 2003). The
56
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
Twangiza trend (Figure 7-8a) is a north easterly interpreted structural corridor which controlled different
gold deposits in Congo. The Geita trend (Figure 7-8a) is a north west trending interpreted lineament that
control major gold trends in Tanzania. These structural corridors have showed consistency with the major
regional trends like Aswan shear zone and the western rift valley. These trends are continued further to
the north affecting different areas in SW of Uganda including Buhweju (Figure 7-8a).
57
A)
Figure 7-6: Field data and lithological interpretation on mineralization. A) Interpreted lithological units and gold
occurrence areas. Schist and Mudstone are the common lithologies which hosted most of the gold fields in Buhweju.
B) Field photo from Kitaka mine. Sulphide vein found disseminated throughout the schist (host rock) C) Field photo
showing typical morphology of alluvia deposit in Buhweju. Always located downstream and covered with vegetation.
In Buhweju, three common structural orientations were clearly recognized. Two of the orientations have
showed consistency with the trend of the Twangiza and Geita structural corridors. In the study area
different magnetic faults were interpreted (Figure 7-5) among which the most prominent fault zone
identified trends in NNE-SSE direction which could be the extension of the Twangiza structural corridor.
This fault is characterized by high lineament density (Figure 7-7a) and high subsurface magnetic anomaly.
In addition most of the gold occurrences in the study area are spatially controlled by this fault (Figure 7-
9a). The NNW-SSE trending structure in Buhweju could also be interpreted in terms of the Gieta trend
and the regional Aswan shear zone. The high lineament density zone and densely clustered gold
occurrences at Katonga might be due to the intersection of these oppositely trending structures. Generally
two trends could be recognized (Figure 7-9b) from the distribution of gold occurrences of the area that
could be related with the locally identified structures as well as the regional structural trends.
58
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
Figure 7-7: Distribution of gold occurrence areas. A) The relationship between lineament density
and gold occurrences. Circles indicate the two clustered areas of gold occurrences and their density
pattern. B) Lithology and mineral occurrence association. C) Subsurface magnetic signature
beneath mineral occurrence areas. D) Quartz veins and gold occurrences. Explanation is given in
detail in the text.
59
Figure 7-8: The Twangiza and Gieta trends A) The gold occurrences in Congo and Tanzania controlled by the
Twangiza and Gieta gold trends (source: (Corporation, 2006; Magnus, 2003) B) The relationship established
between the gold trends and orientations of lineaments interpreted in the area. (a) Shows rose diagrams for surface
lineaments (b) for sub surface lineaments (c) for quartz veins (d) trends of high magnetic anomaly bodies
Figure 7-9: Structural corridors A) the spatial association of gold occurrences with interpreted
magnetic faults B) General alignment of gold occurrences in the area which showed consistency with
the interpreted lineament orientation as well as the Twangiza and Geita trends
60
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
Even though it was difficult to quantify how useful information was provided by each data set, in this
study all the available data sets have provided useful information. Particularly the gamma ray data has
found to be an invaluable aid, which together with modern enhancement and integration techniques have
assisted in the visualization and discrimination of lithological units of the area. Among the various image
enhancement techniques image fusion and pan-sharpening of the radiometric composite image with
multispectral data and SRTM DEM have resulted better lithological differentiation in the area. On the
basis of similar radiometric signature different Precambrian basement lithologies including schist, gneiss
and granite could be identified and mapped which were not mapped before. The SRTM DEM of the area
has revealed topographic and structural information of the complex Buhweju plateau. Useful information
was also depicted from aeromagnetic data about the distribution of sub surface magnetic domains.
Subsequent extractions of both surface and subsurface lineaments were carried on multispectral, SRTM
DEM and aeromagnetic data. The orientation analysis has revealed three dominant lineament orientations
which determine the tectonic grain of the area, NNW-SSE, ESE-WNW and NNE-SSW. These
orientations were found consistent with the two major regional tectonic trends. The NNE-SSW has
considered a major tectonic grain of the area related to the western rift valley while the NNW-SSE seemed
attributed to the regional Aswan shear zone. Non-orientation density analysis applied on the individual
surface and subsurface lineaments has indicated that the Buhweju group rocks were highly affected by
surface lineamnents whereas the Precambrian basement rocks were affected mainly by subsurface
lineaments. Generally the density patterns of surface and subsurface lineaments are weakly correlated.
61
8.2. Recommendations
Various enhancements applied in this study resulted better images for visual interpretation of structures
and lithological units. However, two main factors are identified which greatly hinder the interpretability of
remote sensing data especially multispectral and gamma ray. These are topographic effect and vegetation
cover. More than half of Buhweju is covered by two dense forests protected by forestry department of
Uganda. Moreover, Buhweju plateau is considered to be the highest land in west of Uganda. These
factors have seriously affected the information content of both multispectral and gamma ray data. This
study has made an effort to suppress the effect of vegetation even though the effect was still significant.
Therefore, this study highly recommend for the next studies to be conducted on Buhweju based on
remote sensing tools to seriously consider the effect of vegetation. Similarly the topography effect on
gamma ray data is significant by affecting the distribution of radioelement around the Buhweju plateau for
which this data should be exclusively corrected for topographic effects of Buhweju.
This study has better defined the tectonic setting of Buhweju area through integrated analysis of surface
and subsurface data. Hence similar approaches are recommended to other areas of SW Uganda. The
source of gold mineralization in Buhweju is still not fully known but this study has at least identified
subsurface high magnetic zones, magnetic faults and surface lineaments which could contribute an input
pertaining to search for source of mineralization. Therefore, detail geochemical investigation and high
resolution geophysical data interpretation are recommended for further understanding of the source of
mineralization in Buhweju gold field. New lithological boundaries are identified from multispectral data
interpretation but have not incorporated in the final map due to inaccessibility of the area but
recommended that they need to be verified.
62
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
LIST OF REFERENCES
Abdullah, A., Akhir, M. J., & Abdullah, I. (2010). Automatic Mapping of Lineaments Using Shaded Relief
Images Derived from Digital Elevation Model (DEMs) in the Maran Sungi Lembing Area,
Malaysia. 15, 949-956.
Allek, K., & Hamoudi, M. (2008). Regional-scale aeromagnetic survey of the south-west of Algeria: A tool
for area selection for diamond exploration. Journal of African Earth Sciences, 50, 6778.
Alsaud, M. M. (2008). Structural mapping from high resolution aeromagnetic data in west central Arabian
Shield, Saudi Arabia using normalized derivatives. 1, 129-136.
Baker, B. H. (1971 ). Explanatory note on the structure of the southern part of the African Rift System
Tectonics of Africa 6 543-548
Benomar, T., & Fuling, B. (2005). Improved geological mapping using landsat-5 TM data in Weixi area,
Yunnan province China. Geo-Spatial Information Science, 8(2), 110-114.
Black, R. (I967 ). Chronique Mines. 364, 225.
Buchwaldt, R., Toulkeridis, T., Todt, W., & Ucakuwun, E. K. ( 2008). Crustal age domains in the Kibaran
belt of SW-Uganda:Combined zircon geochronology and Sm-Nd isotopic investigation
African Earth Sciences, 51, 4-20.
Butt, C. R. M., Lintern, M. J., & Anand, R. (2000). Evolution of regolith and landscapes in deeply
weathered terrain- Implication for geochemical exploration. Ore Geology Reviews, 16(3-4), 168-183.
Cahen, L., Snelling, N. J., Delhal, J., & Vail, J. R. ( 1984). The geochronology and evolution of Africa. .
Casas, A. M., Cortes, A. L., Maestro, A., Soriano, M. A., Riaguas, A., & Bernal, J. (2000). A program for
lineament length and density analysis. Computers and Geosciences 26, 1011-1022.
Chavez, P. S., Sides, S. C., & Anderson, J. A. (1991). Comparison of three different methods to merge
multi-resolution and multi-spectral data: LANDSAT TM and SPOT Panchromatic. Photogrammetic
Engineering and Remote Sensing, 57, 295303.
Chavez, S. P., Sides, C. S., & Anderson, A. J. (1991). Comparison of 3 different methods to merge
multiresolution and multispectral data-Landsat TM and Spot Panchromatic. Photogrammetric
Engineering and remote sensing, 57, 295-303.
Chernicoff, C. J., Richards, J. P., & Zappettini, E. O. (2002). Crustal lineament control on magmatism and
mineralization in northwestern Argentina: geological, geophysical, and remote sensing evidence.
Ore Geology Reviews, 21(3-4), 127-155.
Cobme, A. D. (1932). The geology of Southwest Ankole and adjucent territories with special reference to tin deposit.
Corporation, B. (2006, 04/02). 2006 Exploration Program. Retrieved 13/02, 2011, from
http://www.banro.com/s/Twangiza.asp?ReportID=307246&_Title=2006-Exploration-Program
Crippen, E. R., & Blom, G. R. (2001). Unveiling the Lithology of Vegetated Terrains in Remotely Sensed
Imagery. Photogrammetric Engineering & Remote Sensing, 67, 935-943.
Dickson, B. L., & Scott, K. M. (1997). Interpretation of aerial gamma ray surveys-adding the geochemical
factors. Journal of Australian geology and geophysics, 17(2), 187-200.
Drury, S. A. (2001). Image interpretation in geology: Nelson Thornes Blackwell Science.
Duval, J. S. (1983). Composite color images of aerial gamma-ray spectrometric data. Geophysics, 48(6), 722-
735.
Elliot, G. F. S., & Gregory, J. W. (1895). The geology of Mount Ruwenzori and some adjoining regions of
equatorial Africa. . 51, 669680.
Gabert, G. (1990). Lithostratigraphic and tectonic setting of gold mineralization in the Archean cratons of
Tanzania and Uganda, East Africa. Precambrian Research, 46(1-2), 59-69.
Gad, S., & Kusky, T. (2007). ASTER spectral ratioing for lithological mapping in the Arabian-Nubian
shield, the Neoproterozoic Wadi Kid area, Sinai, Egypt. Gondwana Research, 11(3), 326-335.
Gay, S. P. (1972). Fundamental characterstics of aeromagnetic lineaments:thier geological significances and
their gnificance to geology. Salt Lake City, Utah.
Ghazala, H. H. (1993). Geological and structural interpretation of airborne surveys and its significance for
mineralization, South Eastern Desert, Egypt. Journal of African Earth Sciences (and the Middle East),
16(3), 273-285.
Gillespie, A., Kahle, A., & Walker, E. R. (1987). Color enhancement of highly correlated images. II.
Channel ratio and chromaticity transformation techniques. Remote Sensing of Environment, 22, 343-
365.
63
Grant, F. S. (1985). Aeromagnetics, geology and ore environments. Magnetite in igneous, sedimentary and
metamorphic rocks : An overview. Geoexploration, 23, 303-333.
Grant, N. K. (1967 ). Nature. 215, 609.
Grant, N. K. (1969 ). Bulletin of Geological Society. 80, 45.
Groves, D. I., Goldfarb, R. J., Gebre-Mariam, M., Hagemann, S. G., & and Robert, F. (1998). Orogenic
gold deposits: A proposed classification in the context of their crustal distribution and
relationship to other gold deposit types. Ore Geology Reviews, 13, 7-27.
Gunn, P. J., Maidment, D., & Milligan, P. R. (1997). Interpreting aeromagnetic data in areas of limited
outcrop. 17(2), 175-185.
Gunn, P. J., Minty, B. R. S., & Milligan, P. R. (1997). The Airborne Gamma-Ray Spectrometric Response
Over Arid Australian Terranes. Radiometric Methods and Remote Sensing, 96, 733-740.
Gupta, R. P. (1991). Remote Sensing Geology Berlin: Heidelberg: Springer-Verlag.
Harris, J. F. (1961). Summary of the geology of Tanganyika. Economic geology, 1.
Henderson, B. D., Ferrill, A. D., & Clarke, C. K. (1996). Mapping geological falukts using image
processing techniques applied to hill-shaded digital elevation models. 240-245.
Hepworth, J. V., & Macdonald, R. (1966). Orogenic belts of the northern Uganda basement., 210, 726
727.
Ikingura, J. R., Reynolds, P. H., Watkinson, D. H., & Bell, K. (1992). 40Ar/39Ar dating of micas from
granites of NE Kibaran Belt (Karagwe-Ankolean), NW Tanzania. Journal of African Earth Sciences
(and the Middle East), 15(3-4), 501-511.
Jaques, A. L., Wellman, P., Whitaker, A., & Wyborn, D. (1997). High resolution geophysics in modern
geological mapping. 17(2), 159-173.
Jayawardhana, M. P., & Sheard, N. S. (2000). The use of airborne gamma-ray spectrometry- A case study
from the Mount Isa Inlier, Northwest Queensland, Australia. Geophysics, 65, 1993-2000.
Johnson, S. S., Gathright, M., T., & Henika, S. W. (1979). Gamma-ray Specrtometry and Geologic
mapping. Remote Sensing of Environment, 25, 17-24.
Khan, S. D., & Glenn, N. F. (2006). New strike-slip faults and litho-units mapped in Chitral (N. Pakistan)
using field and ASTER data yield regionally significant results. International Journal of Remote Sensing,
27(20), 4495 - 4512.
Klerkx, J., Liegeois, J. P., Lavreau, J., & Claessens, W. (1987). Crustal evolution of the northern Kibaran
Belt, eastern and central Africa. 217- 233.
Leggo, P. J. (1974). A geochronological study of the basement complex of Uganda., 130, 263277.
Link, K., Koehn, D., Barth, M., Tiberindwa, J., Barifaijo, E., Aanyu, K., et al. (2010). Continuous cratonic
crust between the Congo and Tanzania blocks in western Uganda. International Journal of Earth
Sciences, 1-15.
MacDonald, R. ( 1966). Uganda Geology, Scale 1:1,500,000.- Department of Geological Survey Mines;
Entebbe.
Magnus. (2003, 12/2009). Uganda Projects. Retrieved 13/02, 2011, from
http://www.magnusresources.com/africa_uganda_projects.htm
Masoud, A., & Koike, K. (2006). Tectonic architecture through Landsat-7 ETM+/SRTM DEM- derived
lineaments and relationship to the hydrogeologic setting in Siwa region, NW Egypt. Journal of
African Earth Sciences 45, 467-477.
Muwanga, A., Schumann, A., & Biryabarema, M. ( 2001). Landslides in Uganda, Documentation of a
Natural Hazard., 136, 111-115.
Nicolet, J. P., & Edric-Krausz, G. (2003). Guidelines for radioelement mapping using gamma ray spectrometry data.
Vienna, Austria: IAEA.
OLeary, D. W., Friedman, J. D., & Pohn, H. A. (1976). Lineament, linear, lineation: Some proposed new
standards for old terms. Geological Society America Bulletin, 87, 1463-1469.
Owusu, A. E., Woldai, T., Carranza, M. J. E., & Hale, M. (2006). Analysis and Integration of Landsat
ETM and SRTM Data for Geological Mapping of the Kibi Area in Ghana. 330-338.
Owusu, E. A., Woldai, T., Carranza, E. J. M., & Hale, M. (2006). Analysis and Integration of Landsat
ETM and SRTM Data for Geological Mapping of the Kibi Area in Ghana.
Perrotta, M. M., Almeidab, R. I. T., Andrade, F. B. J., Souza Filhoc, R. C., Rizzotto, J. G., & Santosb, M.
G. M. (2008). Remote Sensing Geobotany and Airborne Gamma-ray data Applied to Geological
Mapping within Terra Firme Brazilian Amazone Forest: A Comparative study in the Guapore
Valley (Mato Grosso State, Brazil). Remote Sensing and Spatial Information Sciences
275-1280.
64
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
Pohl, W. (1994). Metallogeny of the northeastern Kibaran belt, Cental Africa-Recent Perspectives. Ore
Geology Reviews, 9, 105-130.
Porwal, A., Carranza, E. J. M., & Hale, M. (2006). Tectonostratigraphy and base-metal mineralization
controls, Aravalli province (western India): New interpretations from geophysical data analysis.
Ore Geology Reviews, 29(3-4), 287-306.
Qaid, M. A., & Basavarajappa, T. H. (2008). Application of Optimum Index Factor Technique to Landsat-
7 Data for Geological Mapping of North East of Hajjah, Yemen. 1 84-91.
Qari, M. Y. H. T. (1992). Improving geological mapping in the Arabian Shield using Landsat Thematic
Mapper imageryJabal Duhayyah area. International Journal of Remote Sensing, 13(15), 2765 - 2772.
Quadros, P. F. T., Koppe, C. J., Strieder, J. A., & Costa, L. C. F. (2003). Gamma-Ray Data Processing and
Integration for Lode-Au Deposits Exploration. Natural Resources Research 12, 57-65.
Radulescu, J. (1982). Mineralization in the Karagwe-Ankolean system of East African/Burundi: In the
development potential of precambrian mineral deposits. 217-225.
Ramadan, M. T., El Mongy, A. S., & El Dein, S. S. (2009). Exploration for Uranium and Thorium
Mineralizations at Wadi Um Laseifa Area, Central Eastern Desert, Egypt: Using Remote Sensing
Technique. Australian Journal of Basic and Applied Sciences, 3(2), 689-697
Reece, A. W. (1961). Explanation of the Geology of sheet 76 (Buhwezu).
Riad, S., & El Etr, H. A. (1985). Bouguer Anomalies and Lithosphere-Crustal Thickness in Uganda Journal
of Geodynamics, 3, 169-186.
Rowan, L. C., & Mars, J. C. (2003). Lithologic mapping in the Mountain Pass, California area using
Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) data. Remote
Sensing of Environment, 84(3), 350-366.
Rumvegeri, B. T. (1991). Tectonic significance of Kibaran structures in Central and Eastern Africa. Journal
of African Earth Sciences (and the Middle East), 13(2), 267-276.
Saggerson, E. P., & Turner, L. M. (1972 ). 24th International Geological Congress 1, 153.
Saraf, A. K., Mishra, P., Mitra, S., Sarma, B., & Mukhopadhyay, D. K. (2002). Remote sensing and GIS
technologies for improvements in geological structures interpretation and mapping. International
Journal of Remote Sensing, 23(13), 2527 - 2536.
Sarapirome, S., Surinkum, A., & Saksutthipong, P. (2002). Application of DEM Data to Geological
Interpretation: Thong Pha Phum Area, Thailand.
Schetselaar, E. M., Chung, C. J. F., & Kim, E. (2000). Integration of Landsat TM, Gamma-Ray, Magnetic,
and Field Data to Discriminate Lithological Units in VegetatedGranite-Gneiss Terrain. Remote
Sensing of Environment, 71(1), 89-105.
Schetselaar, E. M., & Ryan, J. J. ( 2009). Remote predictive mapping of the Boothia mainland area,
Nunavut, Canada: an iterative approache using landsat ETM, aeromagnetic and geological field
data. Canadian Journal of Remote Sensing, 35, S72-S94.
Schetselaar, E. M., Tiainen, M., & Woldai, T. (2008). Integrated geological interpretation of remotely
sensed data to support geological mapping in Mozambique. 48, 35-63.
Shackleton, R. M. (1976). Pan-African Structures. 280, 491-497.
Shives, R. B. K., Charbonneau, B. W., & Ford, K. L. (2000). The detection of potassic alteration by
gamma-ray spectrometry-Recognition of alteration related to mineralization. 65(6), 2001-2011.
Singh, B., & Dowerah, J. (2010). ASTER DEM Based Studies for Geological Investigation around
Singhbhum Shear Zone (SSZ) in Jharkhand, India. Journal of Geographic Information System, 2, 11-14.
Soulakellis, A. N., Novak, D. I., Zouros, N., Lowman, P., & Yates, J. (2006). Fusing Landsat-5/TM
Imagery and Shaded Relief Maps in Tectonic and Geomorphic Mapping: Lesvos Island, Greece.
Photogrammetric Engineering & Remote Sensing, 72,, 693700.
Tack, L., Duchesne, J. C., Lirgois, J. P., & Deblond, A. (1994). Two successive mantle-derived A-type
granitoids in Burundi: Kibaran late-orogenic extensional collapse and lateral shear along the edge
of the Tanzanian craton.
Tanner. (1970). The Ruwenzori Fold Belt of East Africa. 14, 37.
Tsige, L. (2006). Metamorphism and gold mineralization of the KentichaKatawicha area: Adola belt,
southern Ethiopia. Journal of African Earth Sciences, 45 1632.
Wayland, E. J. (1934). crystalline gold and the possibilities of auriferous lodes in the vicincity of the Muti stream, in
Kenya-Uganda minerals exploration.
Wickert, L. M., & Budkewitsch, P. (2004). ASTER - a geological mapping tool for Canada's North. Case
Study: The Belcher Islands, Hudson Bay, Nunavut, Canada. 2 1300 - 1303
Wilford, J. R. (2002). Airborne Gamma-ray Spectrometry. Geophysical and Remote Sensing Methods for Regolith
Exploration, 46-52.
65
Wilford, J. R., Bierwirth, P. N., & Craig, M. A. (1997). Application of airborne gamma-ray spectrometry in
soil/regolith mapping and applied geomorphology. 17(2), 201-216.
Williams, R. S. (1983). Manual of Remote Sensing American Society of Photogrammetry, 1667-1951.
Won-In, K., & Charusiri, P. (2003 ). Enhancement of thematic mapper satellite images for geological
mapping of the Cho Dien area, Northern Vietnam. International Journal of Applied Earth Observation
and Geoinformation, 4, 183193.
Yesou, H., Besnus, Y., & Rolet, J. (1993). Extraction of spectral information from Landsat-TM data and
merge with spot panchromatic imagery- Acontribution to the study of geological structures.
photogrammetry and remote sensing, 48, 23-36.
ZHANG, Y., XIONG, S., & CHEN, T. (1998). Application of Airborne Gamma-ray Spectrometry to
Geoscience in China. Applied Geophysics, 49(2 ), 139-146.
66
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
Appendix 1
Field collected lithological and structural data
ID X Y Z Litho-unit strike_dip dip dire/amount mine structutre
1 216758 9981046 1442 schist mine
2 211492 9978918 1524 micacious schist 40w/20 220/20
3 211252 9979763 1589 Fe rich quarzite
4 211107 9980183 1618 quarzitic conglomerate
5 211101 9980147 1626 conglomerate 10E/40 290/40
6 211076 9980198 1625 folded coglomerate Anticline
7 210598 9980288 1686 bedded sandstone
8 209923 9976519 1865 quarzite vein
9 209272 9968854 1865 quarzite qtz vein
10 208588 9966327 1848 quarzite gold minevein
11 208226 9965874 1848 quarzite mine vein
12 205130 9964770 1792 quarzite mine
13 202269 9963759 1639 granitic gniess
14 201356 9965460 1608 pellite 70w/20 250/20 vein
15 200377 9985882 1152 dolerite N70E mine qtz vein
16 200344 9985831 1182 dolerite kitaka
17 203148 9988118 1375 mica schist 080/90 350/90
18 203508 9988969 1334 amphibolite
19 204237 9987834 1381 schist
20 204342 9986184 1322 gniess 158/60 68/60
21 204242 9986223 1306 mica schist
22 204242 9986576 1312 gniess
23 198261 9989403 1057 gniess
24 194642 9988410 987 gravel, sand cilt
25 192263 99888028 955 gravel, sand cilt
26 190907 9985451 984 gravel, sand cilt
27 192463 9980082 1112 amphibolite 030/44 120/44
28 190203 997852 1068 conglomerate
29 190609 9978263 1138 quarzite 044/50 134/50
30 187216 9978479 1106 tuff
31 182042 9977883 1069 tuff syncline
32 177598 9977190 1075 tuff
33 176005 9972510 1248 tuff anticline
34 177576 9967524 1357 tuff
35 176367 9963119 1484 quarzite
36 176393 9961286 1568 quarzite 170/40 080/40
37 176630 9961047 1593 schist+quarzite 164/36 74/36
38 176573 9961085 1587 quarzite 070/20 160/20 syncline
39 181416 9956287 1408 gniess 040/58 130/58 mine
40 182084 9956592 1401 schist mine
67
41 182785 9957714 1424 schist mine
42 182831 9957757 1427 schist mine vein
43 183032 9958246 1461 schist vein
44 195495 9986150 994 quarzite
45 196251 9985397 1007 quarzite
46 196037 9984917 927 quarzite+schist
47 197203 9984437 1048 quarzite+schist
48 197343 9983817 1097 quarzite
49 197544 9983717 1125 biotite schist
50 197589 9983660 1137 biotite schist
51 202263 9982348 1378 schist mine vein
52 198312 9982877 1244 schist mine vein
53 196182 9983524 985 amphibolite 198/40 108/40
54 196242 9983381 977 granite mine vein
55 196174 9983303 981 granite
56 196274 9983118 964 granite mine
57 204180 9981516 1394 granite vein
58 205899 9985230 1510 schist 90/76 vein
59 205746 9985342 1471 schist
60 207298 9988723 1405 schist N30E
61 205383 9986749 1449 schist
62 205050 9982709 1448 schist
63 207783 9942882 1471 granite 100/70 010/70 vein
64 209107 9944694 1536 granite
65 208803 9946156 1572 granite vein
66 208902 9948392 1581 quarzite
67 215842 9953441 1525 schist
68 214810 9954058 1493 swampy area
69 214188 9954679 1505 conglomerate
70 213735 9956183 1525 schist
71 213884 9957293 1558 schist
72 213223 9958826 1547 conglomerate
73 214314 9960017 1588 quarzite
74 211815 9960615 1592 schist
75 210826 9960248 1672 schist
76 209003 9959668 1723 schist
77 212274 9960417 1572 schist mine
78 215649 9958750 1639 schist
79 216437 9957996 1723 schist
80 216741 9957333 1597 schist
68
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
69
121 205335 9970873 1467 schist
122 205391 9969870 1444 clay
123 204908 9969678 1407 clay_quarzite_cement mine_cement
124 204848 9969677 1392 quarzite
125 204528 9968734 1474 quarzite_mudstone
126 204570 9967809 1495 conglomerate
127 204359 9967757 1495 mudstone N50E mine
128 204627 9967211 1514 quarzite mine
129 204893 9966970 1534 quarzite N30W mine
130 204966 9966961 1538 quazite mine
131 204257 9967653 1501 quazrzite N20W mine
132 202754 9967313 1528 slate 170/10 260/10
133 202571 9966278 1517 schist?/ mine
134 204168 9962951 1699 quarzite mine
135 204616 9961614 1734 schist
136 206155 9958798 1736 schist
137 205291 9957320 1843 schist
138 203342 9955232 1928 schist
139 201398 9954714 1945 schist
140 201479 9952202 1785 quarzite
141 179462 9960225 1400 schist
142 179536 9959894 1490 schist_quarzite
143 180965 9959756 1413 schist
144 182472 9960197 1385 schist
145 182614 9960354 1364 schist
70
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
Appendix 2
Description and Radiometric signatures of the major lithological units of Buhweju
71
Appendix 3
Confusion matrix accuracy report
72
INTER-RELATIONSHIP BETWEEN LITHOLOGY AND STRUCTURE AND ITS CONTROL ON GOLD MINERALIZATION IN BUHWEJU AREA, SW OF UGANDA
73