An integrated framework for global detection and quantification of Lunar Pyroclastic deposits

Lunar pyroclastic deposits (LPDs) composed of Fe–Ti-bearing volcanic glasses are considered one of the best proxies to understand the primitive lunar mantle composition. These deposits are characterized by very low albedo and smooth surfaces associated with volcanic landforms. While regional LPDs can be identified based on these features, the approach is inadequate for detecting localized LPDs. The spectral signatures of volcanic glasses overlap with the common lunar minerals in the visible-to-near-infrared wavelength range, also making their detection challenging through remote observations. In this work, we propose a framework focusing on systematic identification of LPDs by incorporating morphological context with a newly derived spectral-parameter-based false-color-composite scheme using Moon Mineralogy Mapper (M3) data. Additionally, for the first time, we applied a probabilistic hyperspectral unmixing approach for remote estimation of volcanic glass abundances within LPDs. We evaluated this integrated framework on the well-known regional LPD at the Aristarchus crater region. Our result effectively detected the spatial extent of pyroclasts with an estimation of ∼75 wt% volcanic glasses across the identified deposit. We applied this approach to the M3 global mosaic (∼1.5 km pixel−1) and successfully detected several known localized and regional LPDs. Out of those, five locations were mentioned with their respective volcanic glass concentrations. The developed framework can lead to a comprehensive global identification and detailed quantitative analysis for characterization and classification of the LPDs. This approach will advance our understanding of the complex nature of lunar explosive volcanism and the geological evolution of the Moon.