Figure
1: Shematic map of West Africa Craton (Boher et al., 1992)
Figure 2 . Presentation of the study area on geological map of
WAXI (2018)
2. MATERIAL
The equipment used to carry out this work includes remote analysis data
and field data. Remote analysis data includes satellite image Landsat 8
and the radarsat image. In addition to these images, there is the
geological map of WAXI 2018. The images were downloaded and processed
from softwares. These are Landsat 8 optical image, radarsat image of the
study area. Satellite images Landsat 8 were downloaded from the website
at http://www.earthexplorer.usgs.gov. According to Landsat image, we
used a total of 11 images (bands 1, 2, 3, 4, 5, 6-1, 6, 7, 8, 9, 10 and
11). Aeromagnetic image has been treated also.The survey was carried out
by Kenting Earth Sciences Ltd, during the years 1974 and 1975 as part of
a cooperation program between the Government of Canada and Côte
d’Ivoire.The softwares used for image processing are Envi, Arc Gis,
Google Earth Pro and Qgis, Geosoft. Structural data was processed with
Global Mapper, Georient and Leapfrog softwares.
3. METHODS
Relative to the landsat 8 image, the band 8 has been choosen. The band 8
is the panchromatic band with a resolution of 15 meters on the ground.
After atmospheric and radiometric corrections applied to band 8,
Principal Composition Analysis (PCA) and various processing filters were
applied to enhance the perception of linear structures. PCA is an
effective technique which allows to accentuate multi-spectral images for
fine geological interpretations (Biémi et al., 1991) because it allows
to reduce the information contained in several bands, sometimes highly
correlated (redundancy of spectral information) in a number more
restricted of components. Component 1 (band 1 of PCA) was retained.
Therefore, we have chosen to apply the filter to this first principal
component. Filters have also been applied to the radarsat image.
Filtering consists in changing the value of a pixel according to those
of its neighbors (Touzi et al., 1988, Nezry, et al., 1991; Lopez, et
al., 1992; Yésou, et al., 1993). We applied the four Sobel filters, the
PREWITT filter and the YESOU filter to the two images. These directional
filters whose application matrices make it possible to bring out or mask
specific characteristics of an image based on their frequency related to
texture (Himyari, et al., 2002; Jourda, et al., 2006; Ta, et al,. 2008;
Djemai, et al., 2009; Guergour, et al., 2009; Kouamé, et al., 2009).
Then, the flowchart of the processing carried out on the images is
presented in figure 3. The lineament maps resulting from these images
were obtained. Finally, we validated the Landsat image and the radarsat
image using the field data and pre-existing maps. After extracting the
study area from the raw aeromagnetic map, the processing consists of
applying specific filters. We successively applied the horizontal and
vertical derivatives and the magnitude of the horizontal gradient. The
calculation of the Euler deconvolution, by applying the structural index
”1”, made it possible to highlight faults at depth and to follow their
connection with the faults affecting the cover in accordance with the
work of ( Vanié and al., 2005, Vanié et and., 2006, El Gout et al.,
2009, Khattach, 2010). The results of the various treatments are
presented as lineament maps and rose diagram. (Vanié et al., 2006).
Figure 3.a Landsat8 Image processing flowchart; bRadarsat Image processing flowchart; c AeromagImage processing
flowchart
During the differents geological mapping campaigns, the rare outcrops
that appear in contact with the granite Ferkessedougou batholith and the
metasediments, along the watercourses and finally along the tracks
allowed us to identify several structural elements. Although some
outcrops are not directly located on the mineralization, nevertheless
they offer several east-west sections through the granite contact
(Ferkessedougou batholith ) and the metasediments. Also, it will be
necessary to add the cartography of 7100m in length of thirty-seven (37)
trenches during which nine hundred eighty-seven (987) structures were
measured and one thousand nine hundred and sixty (1960) structures from
nine ( 19) core drilling. The planar structures are bedding, shearing
foliation, veins or sometimes quartz and carbonates veinlets. All of
these data have been classified according to their nature and their
chronological relationship, in to deformation structures D1, D2 and D3.