<p>The Upper Proterozoic Bou Azzer ophiolite complex is a globally significant metallogenic province in the central Anti-Atlas region of Morocco, particularly due to the Co-Cr-Ni mineralization in a structurally complex Pan-African geodynamic setting. This study presents an advanced approach to mineral exploration mapping in the Aït Ahmane area by integrating remote sensing, geochemistry, and spatial analysis techniques. Multispectral data from the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) were exclusively used in this study and processed using Principal Component Analysis (PCA) and band ratios to enhance the spectral signatures of key lithological units and hydrothermal alteration minerals, including carbonates, silicification, iron oxides/hydroxides and mafic rocks of the ophiolitic complex. These remote sensing-signatures were systematically combined with cobalt-chrome-nickel geochemical anomalies of on one hand and Copper mineralization anomalies on the other. A data-driven approach was employed to develop a predictive mineralization model based on logistic regression. The choice of this model is justified by its ability to estimate probabilities, aligning with the objective of generating a mineral prospectivity map grounded in probabilistic values. The predictive performance of the model was evaluated using statistical indicators such as the Kappa coefficient and overall accuracy, ensuring the robustness of the results obtained. Results highlight a strong spatial correlation between mineralized zones and fault-controlled ultramafic units, which was successfully highlighted using ASTER VNIR-SWIR and TIR data, underscoring the tectonic control on hydrothermal alterations and ore deposition, along ophiolite boundaries. This integrated methodology provides a robust mineral exploration framework and improves the prospection of Co-Cr-Ni mineralization in ophiolite environments like Aït-Ahmane. The results highlight the distinct deposition processes between Cupriferous and Co-Cr-Ni mineralizations, highlighted with their different sensitivity to remotely sensed data proxies. Furthermore, the study underscores the effectiveness of ASTER data, statistical analysis, and geospatial modeling in optimizing exploration targets in semi-arid and remote similar areas.</p>