Satellite images are widely used in various fields of geoscience, such as geology, for distinguishing different lithological units, delineating regional and local structures such as fractures and faults, and targeting mineral deposits.

Phlogopite is one of the most exploited minerals by the mining industry in Madagascar. The majority of these deposits are located in the southern part of the country, mainly in the Bekily district. Its exploitation dates back to the pre-colonial era and continues to this day, primarily through artisanal or semi-industrial methods. However, these extraction techniques lead to inefficient exploitation of the deposits, resulting in resource wastage and limited benefits for the local population (Rakotondrabe, 2021).

Geologically, southern Madagascar is bounded to the north by the major left-lateral Ranotsara shear zone. This block is highly complex, characterized by various deformations, including tight and reclined folds that dominate the northern part (Noizet, 1971). Beraketa belongs to the Androyen domain, equivalent to the Bekily Block (Beger et al., 2014), and is located at the center of the southern block (Martelat et al., 2000).

This study aims to identify phlogopite mining areas through the analysis of ASTER L1T images. The data used and processed include visible, near-infrared, and shortwave infrared reflectance. The objective is to demonstrate the capabilities of ASTER multispectral optical data (VNIR + SWIR) in terms of mineral indices and spectral analysis for lithological mapping and the delineation of mineralized zones on a large scale within the study area. The VNIR and SWIR bands were normalized using the flat-field calibration method. The digital processing focuses on image enhancement through principal component analysis (PCA) and band ratio calculations (Rakotondrabe, 2020).

The results of satellite image analyses, correlated with geochemical data, fieldwork findings, as well as historical geological maps and mineral indices, reveal a strong correlation with the distribution of phlogopite in the study area. The interpolation of lithological data and mineralization indices shows a consistency in the obtained results. These findings provide a new layer of valuable information that can be used to detect favorable lineaments for more efficient mining operations.