The remote sensing process needs an energy source to illuminate the target; this energy gives off radiation during its pass and interacts with the atmosphere. A second interaction occurs during the path between the target and the sensor. Once the energy is diffused or emitted by the target, it must be captured remotely (by a sensor that is not in contact with the target) to be finally recorded by the sensor and transmitted, often by electronic means, to a receiving station where the information is transformed into images (Gasmi et al., 2016; Mas, 2000).

A visual and/or digital interpretation of the image after processing is necessary to extract desired information to understand the target and help solve a particular problem.

In this work we used the Landsat OLI 8 image, acquired free of charge from the website. (https://earthexplorer.usgs.gov) on 09/11/2021. This satellite image covers a large part of the structures of the Eastern High Atlas. It is an image formed by 11 bands with different wavelengths and special resolutions (Roy et al., 2014) (Table 1).

As well, we used the geological map of the High Atlas of Anoual-Bou Anane (Eastern High Atlas) at 1/200000 (Michard, 1976) Bni Tajjit at 1/50000 (Notes Mém 601.Serv. Géol. Maroc. 2020)covering the study area to ensure the validation of the acquired data by the adopted methodology; where the obtained results are compared with field observations and then validated. Envi, Arc Gis , and RockWorks were used in this satellite image processing study.

Before using the satellite image, we carried out a preprocessing which consists of making a radiometric correction followed by an atmospheric correction and finally applying the Dark “Subtraction (DOS)” technique to eliminate the atmospheric noise and the shadow in the image (Adiri et al., 2017, 2020).

“Enhanced” image processing has been applied to clarify geological features. Techniques include color composites (false color RGB), principal component analysis (PCA), spatial filtering, and reporting band.

The major steps of data preprocessing and processing are summarized in the flow diagram below (Fig. 2).

Several ratio combinations have been used in a number of previous works for the purpose of mapping lithological units. These combinations were tested and applied and validated during this study, two combinations were chosen because of their good discrimination of rock units in the study area using the Landsat 8 OLI image (Table 2).

After the preprocessing of the satellite Landsat 8 OLI image, we made several tests of band combinations in RGB on the basis of the spectral signature and the contrasts of the lithological units. Also, the composite color 257 RGB; PC1-PC4-PC3 (RGB; suggested in this study) was found to be the best method for lithological mapping of the Boudahar massif.

Based on several previous works such as (Choukrad et al., 2020; Tobi et al., 2022), 657 and 257 color composites as RGB images are useful widely in lithologic mapping, they suggest that clay and carbonate minerals are characteristic of absorption of 2.1 to 2.4 µm and reflectance of 1.55 to 1.75 µm. On the same basis, and after visual examination of several RGB band combinations, we found that the 2, 5 and 7 give good discrimination between limestones, sandstones, and alluviums.

The bands (PC2; PC5; PC6) are used for RGB color combinations, which can separate different lithologic facies, this has been verified by field visits and has been confirmed (Fig. 3).

the two images (MNF8 and MNF9) were excluded because the results show that these components did not provide any useful information due to the high level of noise contained in these components. The remaining components are used to produce RGB color composites. The MNF4, MNF7, and MNF2 components are used in RGB. On this image (Fig. 4), most of the lithological units can be identified by simple comparison with the geological map of the area.

Principal component analysis and band ratio techniques have been evaluated in lithologic mapping. In order to have good lithologic outcrop mapping, it is necessary to combine the results of both methods (Fig. 5). The data from the pre-existing map is rechecked by field investigations.

Verification of the acquired maps shows colors that correspond to a series of lithologies that we are trying to find.

Lineaments are defined as structures in lines observed on the Earth’s surface and include natural structures of geological and topographical features as well as anthropogenic features (roads, railways, etc.). In order to have a better identification of structural lineaments such as faults, we proceeded to improve the spatial resolution of the OLI image by applying a panchromatic band which has 15m in spatial resolution (Alonso-Contes, 2011). The spectrum bands were resampled at 15 m using the “Pan Sharpening by Gram-Schmid” method (Adiri et al., 2017, 2020; Mahan & Arfania, 2018).

(Es-Sabbar et al., 2020) In this study, we used visual inspection to find the combination of Ratio and RGB (6/5, 7/6, and 4/7) that showed the best ability to identify characteristics. Then, band 6 (SWIR-1) and PC1 are used to obtain the lineaments from the band 6 and PC1 directional filters in different directions. The result is a set of straight lines that are oriented in the main direction of the Boudahar massif and other transverse zones (Fig. 6).

The validation of these results is done with the Google-Earth image (Fig. 7), and by superimposing lineaments on the true color 4-3-2 image or with data collected in the field. We compare it with the elevation map to eliminate any source of noise (peak). To have a better interpretation of the linear structures and their morphologies, the purpose of this step is to remove the linear-curvilinear anthropogenic characteristics of non-geological origin.