Lithological mapping and automatic lineament extraction using Aster and Gdem data in the Imini-Ounilla-Asfalou district, South High Atlas of Marrakech, Morocco

Abstract

Lithological and lineament mapping using remote sensing is a fundamental step in various geological studies, as it forms the basis for the interpretation and validation of the results obtained. There were two objectives for this study, applied in the Imini-Ounilla-Asfalou district, South High Atlas of Marrakech region: first, lithological mapping by satellite image processing techniques such as ASTER L1B (hight spectral and spatial resolution), namely Principal Component Analysis (PCA), Minimum Noise Fraction (MNF), as well as the application of three types of supervised classification, namely Spectral Angle Mapper (SAM), Maximum Likelihood (ML) and Minimum Distance (MD), on the visible/near-infrared (VNIR) and short-wave infrared (SWIR) spectral bands of our ASTER image; second, an analysis of the distribution of lineaments by automatic extraction using a Global Digital Elevation Model (GDEM) and the PC1 image derived from the PCA transformation applied to the satellite image. The best results are highlighted by the delineation of new facies in relation to the existing map; after confirmation in the field, all of these facies, which include Eocene, Triassic and Jurassic formations, are represented on the new map. The results of lineaments showed that each of them systematically shows a similarity in terms of concentration and orientation, with four preferential oriented systems: NE-SW, E-W, NNE-SSW and NW-SE. The lineaments mainly follow those of the major fault zones, with high concentrations in the northeast and southwest parts of the study area.

A review on advancements in lithological mapping utilizing machine learning algorithms and remote sensing data

“… The combination of ASTER and simulated panchromatic Sentinel-2A data showing the most efficient result [ 85 ] RF/SVM/NB/Classification and regression tree (CART) ASTER-L1T/Landsat-8//Sentinel-2 The utilization of CART for the classification of Landsat-8 data resulted in an impressive accuracy of 99.63%, which showed strong agreement with field validation. [ 86 ] SAM/ML/MD ASTER Ml and MD resulted in improved differentiation of various geological facies. The classification achieved 91.3% accuracy and 90.1% overall precision.…”

Mapping and analysis of structural lineaments using SRTM radar data and Landsat 8-OLI images in Telouet-Tighza area, Marrakech High Atlas - Morocco.

“…Therefore, the automatic extraction method applied on both Landsat 8 OLI and SRTM satellite data showed that these data can be used as a powerful tool to explore structural features and to further improve lineament mapping in several worldwide regions (Jawahar Raj and Prabhakaran 2018) (Aretouyap et al 2020). In Morocco, (Abdelouhed, Ahmed, Abdellah, Mohammed, et al 2021)demonstrated the effectiveness of Landsat 8 OLI and SRTM data to enhance the morphostructural lineaments in Ikniouen area, Eastern Anti Atlas.…”

“…It can detect structural lines with horizontal, vertical, and diagonal directions(Hermi et al 2017). Directional lters were commonly used by several authors in geological applications to enhance and highlight speci c linear trends as well as geological(Hermi et al 2017; Abdelouhed, Ahmed, Abdellah, and Mohammed 2021;Abdelouhed, Algouti, et al 2021). In the present study, the DF technique was selected for the automatic lineament extraction, since it is considered such a faster and effective way to evaluate lineaments and identify any linear or curvilinear shapes (faults, fractures, contours, roads and defects...) on the image(Nedjraoui et al 2021; Azman, Ab Talib, and Sokiman 2020).…”